text
stringlengths 0
7.84M
| meta
dict |
|---|---|
Significant differences in the expected versus observed longevity of implantable cardioverter defibrillators (ICDs).
Implantable cardioverter defibrillator (ICD) is a life-saving therapy for patients at risk of ventricular arrhythmias. Due to its high cost, its cost-effectiveness is highly dependent on its longevity, which is currently only based upon the manufacturer's predicted device life span. We sought to assess the ICDs' longevity and the factors influencing it, and to compare the observed (real life) to the expected (manufacturer's prediction) life span at a device level. We retrospectively identified all patients who underwent an ICD implantation in a tertiary care medical center. For each device, an expected longevity was assigned based on the manufacturer/model, pacing percentage, and number of shocks per year. We defined device failure if the observed survival was shorter than 80 % of the expected. Only devices with follow-up time that exceeded the expected longevity were included. Of the 275 devices in the cohort, 79 (29 %) failed. Median device longevity was 5 years and varied markedly between manufacturers (4.3, 4.8, 5.1, and 6.3 years for Biotronik, St. Jude Medical, Boston Scientific, and Medtronic, respectively). There were significant differences among the manufacturers in device failure rates: 48, 17, 22, and 14 % for Biotronik, St. Jude Medical, Boston Scientific, and Medtronic, respectively). In multivariate analysis, manufacturer, earlier year of implantation, congestive heart failure and chronic renal failure significantly predicted device failure. In conclusion, there is a significant device failure rate among ICDs, with variability among manufacturers, impacting both patients and the medical economic systems.
|
{
"pile_set_name": "PubMed Abstracts"
}
|
Morphological studies on the established cell lines obtained from normal human decidual tissue of an early stage of gestation.
The two cell lines, TTK-1 (E) and TTK-1 (F), were established from normal human decidual tissue of early gestation. The primary culture was initiated by a fragment culture technique in July, 1979 and the cultures were passaged about once every two months. Meanwhile two kinds of the cultures, the epithelial-like cell dominant one and the fibroblast-like cell dominant one, had appeared. The former has been subcultivated and maintained at a constant growth rate and designated as the TTK-1 (E) cell line. The latter showed a gradual decline in growth rate and finally growth ceased at 3 years after the initiation of the culture. The sudden onset of growth and colony formation occurred after 4 months of senescence and the fibroblast-like cell culture has been maintained at a constant growth rate and designated as TTK-1 (F) cell line. Morphological studies revealed that the TTK-1 (E) cell line had epithelial-like characteristics and TTK-1 (F) cell line had fibroblast-like features. Both cell lines showed heteroploid karyotypes and tumorigenicity in nude mice transplantation. The two cell lines appeared to be established by spontaneous neoplastic transformations and should be useful cellular models for the study of malignant endometrial tumors.
|
{
"pile_set_name": "PubMed Abstracts"
}
|
// Python Tools for Visual Studio
// Copyright(c) Microsoft Corporation
// All rights reserved.
//
// Licensed under the Apache License, Version 2.0 (the License); you may not use
// this file except in compliance with the License. You may obtain a copy of the
// License at http://www.apache.org/licenses/LICENSE-2.0
//
// THIS CODE IS PROVIDED ON AN *AS IS* BASIS, WITHOUT WARRANTIES OR CONDITIONS
// OF ANY KIND, EITHER EXPRESS OR IMPLIED, INCLUDING WITHOUT LIMITATION ANY
// IMPLIED WARRANTIES OR CONDITIONS OF TITLE, FITNESS FOR A PARTICULAR PURPOSE,
// MERCHANTABILITY OR NON-INFRINGEMENT.
//
// See the Apache Version 2.0 License for specific language governing
// permissions and limitations under the License.
using System.Reflection;
using System.Runtime.InteropServices;
[assembly: AssemblyTitle("Visual Studio - Python build tasks")]
[assembly: AssemblyDescription("Contains build tasks for Python projects.")]
[assembly: ComVisible(false)]
|
{
"pile_set_name": "Github"
}
|
Sanskrit, which means "refined", "consecrated" and "sanctified," is the classical language of ancient India and it became the liturgical language of Hinduism, Buddhism and Jainism. Recently archaeologists have discovered the earliest epigraphic evidence so far for the Saptamatrika cult in Southern India.
It should be mentioned that in Hinduism, Saptamatrika is a group of seven mother-goddesses, who are the wives of Hindu deities- Brahma, Shiva, Kumara, Vishnu, Varaha, Indra and Yama. However, as per the Epigraphy Branch of the Archaeological Survey of India, this newly found Sanskrit inscription is the earliest evidence to have been found in South India as on date.
The Sanskrit inscription
The ancient evidence was issued by Satavahana dynasty's last king Vijaya that dates back to 207 AD. As mentioned by the Director of Epigraphy branch, Archaeological Survey of India, Dr K Muniratnam, the Sanskrit inscription was unearthed in India's Chebrolu village in Guntur district of Andhra Pradesh earlier this month.
The local residents first found the inscription and informed the authorities. The villagers noticed a pillar with some engravings when they were restoring and repairing a local temple. Muniratnam told Indian news organisation The Hindu that the inscription was first copied and studied by a person named Kartika.
It records the construction of a temple and includes description of a few images on the southern side of the temple. As per Muniratnam, earlier references of Saptamatrika worship were noticed in Kadamba copper plates and the Chalukyas as well as Eastern Chalukya copper plates. But this new finding takes the archaeologists almost 200 years ago which also means that the inscription is earliest datable evidence from South India so far.
Another archaeological finding
From the same archaeological site another inscription, which was written in Prakrit language, was found. It should be mentioned that this inscription belongs to the 1st century AD and is the earliest evidence epigraphic reference to monastery said Dr Muniratnam. He also added that there are many ancient monuments and structures across the country which could contain a treasure trove of information but right now, all of these historic monuments or structures need protection.
|
{
"pile_set_name": "OpenWebText2"
}
|
MANN, CYNTHIA K,
?
You have been selected to participate in the Mid Year 2001 Performance
Management process. Your feedback plays an important role in the process,
and your participation is critical to the success of Enron's Performance
Management goals.
?
To complete a request for feedback, access PEP at http://pep.enron.com and
select Complete Feedback from the Main Menu. You may begin providing
feedback immediately and are requested to have all feedback forms completed
by Friday, May 25, 2001.
?
If you have any questions regarding PEP or your responsibility in the
process, please contact the PEP Help Desk at:
Houston: 1.713.853.4777, Option 4 or email: [email protected]
London: 44.207.783.4040, Option 4 or email: [email protected]
?
Thank you for your participation in this important process.
?
The following is a CUMULATIVE list of employee feedback requests with a
status of "OPEN." Once you have submitted or declined an employee's request
for feedback, their name will no longer appear on this list. NOTE: YOU WILL
RECEIVE THIS MESSAGE EACH TIME YOU ARE SELECTED AS A REVIEWER.
?
Employee Name:
ADAMS, SUZANNE
BERNSTEIN, MARK
CARNAHAN, KATHLEEN
JACOBY, BEN
MAY, THOMAS
MITRO, FRED
RORSCHACH, REAGAN
THOME, STEPHEN
WHITE, ANN ELIZABETH
|
{
"pile_set_name": "Enron Emails"
}
|
Zainadine Júnior
Zainadine Abdula Chavango Júnior (born 24 June 1988 in Mozambique) is a Mozambican football defender currently playing for Marítimo. Previously he has been on trial at Sporting Clube de Portugal, together with fellow Mozambican Mexer; however, it was only Mexer that was given a contract. Instead, Zainadine Júnior, or Fino, has impressed in the Mozambican league Mocambola and for his national team. His performances has earned interest from South African club Jomo Cosmos, but so far no move has materialized for the centre-back.
On 10 January, Zainadine signed with Marítimo until July 2017.
International career
International goals
Scores and results list Mozambique's goal tally first.
References
External links
Category:1988 births
Category:Living people
Category:Sportspeople from Maputo
Category:Mozambican Muslims
Category:Mozambican footballers
Category:Mozambique international footballers
Category:Desportivo Maputo players
Category:Primeira Liga players
Category:C.D. Nacional players
Category:Tianjin Teda F.C. players
Category:Expatriate footballers in China
Category:Chinese Super League players
Category:Association football defenders
Category:Liga Desportiva de Maputo players
|
{
"pile_set_name": "Wikipedia (en)"
}
|
Babichev
Babichev () is a Russian masculine surname, its feminine counterpart is Babicheva. It may refer to
Maxim Babichev (born 1986), Belarusian handball player
Mikhail Babichev (born 1995), Belarusian professional footballer
Roman Babichev (born 1975), Russian football player
Vladislav Babichev (born 1981), Russian volleyball player
Category:Russian-language surnames
|
{
"pile_set_name": "Wikipedia (en)"
}
|
[Candida spondylodiscitis and epidural abscess].
Candida spondylodiscitis associatd with epidural abscess is rarely seen. We present a patient with Hodgkin lymphoma who received chemotherapy and developed systemic Candida infection, which was complicated by Candida spondylodiscitis and epidural abscess.
|
{
"pile_set_name": "PubMed Abstracts"
}
|
Kuragano-shuku
was the twelfth of the sixty-nine stations of the Nakasendō. It is located in the present-day city of Takasaki, Gunma Prefecture, Japan.
History
Kuragano-shuku was an intersection between the Nakasendō and the Nikkō Reiheishi Kaidō. Travelers coming from Kyoto would use this route to get to Nikkō. (If they were coming from Edo, they would have used the Nikkō Kaidō.) During the Edo period, it was a popular port for trader ships on the Karasu River.
Neighboring post towns
Nakasendō
Shinmachi-shuku - Kuragano-shuku - Takasaki-shuku
Nikkō Reiheishi Kaidō
Kuragano-shuku (starting location) - Tamamura-shuku
References
Category:Stations of the Nakasendō
Category:Post stations in Gunma Prefecture
|
{
"pile_set_name": "Wikipedia (en)"
}
|
Field of the Invention
The invention relates to a reset circuit for deactivating a circuit configuration in the event that the supply voltage drops below a specific level to an undervoltage.
A reset circuit of the generic type serves for deactivating an external circuit configuration, e.g. in the form of a microcontroller circuit which is operated by a supply voltage and is intended to be brought to a stable, deactivated state if the supply voltage falls below a specific level, so that logic circuits operated in connection with the microcontroller circuit cannot assume undefined states. In other words, the external circuit configuration is deactivated.
Reset circuits according to the prior art operate, e.g. with a charge pump, which is fed by the supply voltage of the microcontroller, and with a transistor operating as an actuator, across which a bias voltage is dropped. The bias voltage can be kept very small by means of the charge pump during normal operation. In the case of undervoltage supply, the bias voltage assumes a relatively large value and thus reliably deactivates the microcontroller circuit if the supply voltage drops below a specific value. A hysteresis effect is produced in this case, so that the microcontroller circuit xe2x80x9cstarts upxe2x80x9d again only if the supply voltage rises above a specific threshold value.
Circuits of the aforementioned type are relatively complex and expensive.
DE 195 27 603 A1 discloses an electrical circuit configuration for generating a reset signal in the case of undervoltage supply of a microcomputer, which circuit configuration has a series circuitxe2x80x94connected to a voltage to be monitored formed by an npn transistor and a resistor, at the junction point of which the reset signal can be tapped off. The resistor must have a relatively high resistance in order to keep the thermal loading on the transistor within limits. This resistor together with resistors which are located in the microcomputer and are connected to the positive pole of the voltage to be monitored forms a voltage divider, whereby an unambiguous reset signal (low signal) is not ensured.
EP 0 767 416 A1 discloses a circuit for supplying voltage to a microcomputer with generation of a reset signal in the case of undervoltage supply, which circuit has a plurality of capacitors. Undefined states, which should be avoided under all circumstances, can arise in the time between the undershooting of the minimum supply voltage and the appearance of a reset signal.
It is accordingly an object of the invention to provide a reset circuit which overcomes the above-mentioned disadvantageous of the prior art circuits of this general type, and which operates reliably over a wide supply voltage range, requires few components and is thus cost-effective and simple to integrate.
With the foregoing and other objects in view there is provided, in accordance with the invention a reset circuit for, in an event of an undervoltage supply, deactivating a circuit configuration that is fed by a supply voltage. The reset circuit includes two active switching elements having forward paths connected together at a junction point and connected in series between a supply voltage and a ground reference potential. Each one of the two active switching elements has a control terminal. The reset circuit also includes a resistor having one terminal connected to the ground reference potential and another terminal connected to the junction point, at least one forward-biased diode; a first current source connected to the supply voltage via the at least one diode; and a second current source connected to the supply voltage. The control terminal of each one of the two active switching elements is driven by a respective one of the first current source and the second current source. The junction point provides a reset signal.
In accordance with an added feature of the invention, the active switching elements are transistors selected from the group consisting of bipolar transistors and field-effect transistors.
In accordance with an additional feature of the invention, the reset circuit includes a current mirror circuit which includes one of the two active switching elements. The one of the two active switching elements has a drain-source path connected between the supply voltage and the junction point. The one of the two active switching elements is driven by the first current source. The current mirror circuit also includes another active switching element having a drain-source path connected between the first current source and the ground reference potential, the other active switching element being driven by the first current source.
In accordance with another feature of the invention, the reset signal is active low.
In accordance with a further feature of the invention, the reset circuit is used with a motor vehicle electronic circuit.
In accordance with a concomitant feature of the invention, the second current source is directly connected to the supply voltage.
The reset circuit according to the invention has the advantage that a minimum number of components are required and deactivation of the external circuit configuration is performed reliably.
Other features which are considered as characteristic for the invention are set forth in the appended claims.
Although the invention is illustrated and described herein as embodied in a reset circuit, it is nevertheless not intended to be limited to the details shown, since various modifications and structural changes may be made therein without departing from the spirit of the invention and within the scope and range of equivalents of the claims.
The construction and method of operation of the invention, however, together with additional objects and advantages thereof will be best understood from the following description of specific embodiments when read in connection with the accompanying drawings.
|
{
"pile_set_name": "USPTO Backgrounds"
}
|
Reflection | Silent And Strong: Women Lead At Standing Rock
As I march beside hundreds of silent women toward the imposing guards of the Dakota Access Pipeline, I wonder why we don’t do this every damn day, every-damn-where in the world. Arms linked, we march from the camp to the bridge, where behind towering concrete barricades lurks an LMTV (with a gunner!). The smell of sage fills the air, and not a word is uttered. I feel stronger than I ever did during my six years in the army, when an M16 was my useless bodyguard, and anger, my daily drug. Neither anger nor firepower can match the strength of a procession of women peacefully united for a common cause.
The women lead, with the men following behind to “hold the space,” keeping a protectively small distance. Indigenous people, effortlessly confident in the rightness of their action, invite us all to pray with them for the water – and even though praying is not a thing I’ve done in years, I realize the prayer isn’t the kind I performed as a kid in church, but a summoning of the power of the universe to right the wrongs of humanity. We are protecting the water because it is the reason for life, and the women here at Oceti Sakowin Camp know the best way to protect it is not through violent attacks, but through powerful, prayerful, silent presence.
We can tell it’s the women who know because this camp is a matriarchy, and it’s the women who make the decisions – it’s they, not the men, who’ve been calling for the unarmed, peaceful protest that’s been going on here for nearly four months. It’s they who march us forward onto the bridge, then signal: Sit down. We sit. In the front of the march, hands flashing the sign for peace (or perhaps the sign for victory) reach toward the sky, and the gesture ripples back through the crowd. Hundreds of hands, silently floating up toward circling airplanes and surveillance drones, spread their frigid fingers, as the bodies attached to them kneel on the road in the bitter North Dakota wind, praying for the water. We have no idea whether the authorities will attempt to intervene, but we do know that they don’t need a reason to try. We also know that we have every reason to stay exactly where we are.
They don’t intervene. They don’t even make a sound. As daylight begins to fade, we all stand and turn to look over the bridge to our left, where a group of our indigenous leaders are beginning to drum, chant and pray beside the river. Rose-tinted clouds appear in the evening sky, reddening as the sun sinks, melting into blue and then grey as the elders finish their ceremony. A bundle of sage passes through the crowd, its cleansing smoke fanned by hand after hand, and I feel as calm and centered as I ever have, strengthened by the collective intention powering this movement. We all silently link arms and march back to the camp, where an elder thanks us for the work we’ve done today, for completing our action in prayerful silence.
“They didn’t think we could do it, but we did it,” she says, and although I don’t know specifically who “they” are, I understand that this action was done to prove as much to ourselves and our allies as to the Army Corps of Engineers, the government of North Dakota, or the investors in the Dakota Access Pipeline. We’ve shown everyone watching and listening that our quiet, confident presence is our greatest show of strength – not noise, not aggression, not violence, but determination, and peaceful, passionate silence. We are Woman – hear our hearts beat. We don’t need to roar.
|
{
"pile_set_name": "Pile-CC"
}
|
1. Introduction {#sec1-sensors-20-00543}
===============
The capability to estimate where a subject is looking is known as gaze estimation or eye-tracking. This technology has enhanced applications in a wide array of domains, including the measurement of advertising efficacy \[[@B1-sensors-20-00543],[@B2-sensors-20-00543]\], instrumentation to enhance reading \[[@B3-sensors-20-00543],[@B4-sensors-20-00543],[@B5-sensors-20-00543]\], automotive safety \[[@B6-sensors-20-00543],[@B7-sensors-20-00543]\], pilot training \[[@B8-sensors-20-00543],[@B9-sensors-20-00543],[@B10-sensors-20-00543]\], accessibility interfaces \[[@B11-sensors-20-00543],[@B12-sensors-20-00543],[@B13-sensors-20-00543],[@B14-sensors-20-00543]\], and provide objective indicators of cognitive, psychiatric, and neurological states of individuals \[[@B15-sensors-20-00543],[@B16-sensors-20-00543],[@B17-sensors-20-00543],[@B18-sensors-20-00543],[@B19-sensors-20-00543],[@B20-sensors-20-00543],[@B21-sensors-20-00543],[@B22-sensors-20-00543],[@B23-sensors-20-00543],[@B24-sensors-20-00543],[@B25-sensors-20-00543],[@B26-sensors-20-00543]\]. Even though research with specialized eye-tracking systems demonstrated that such systems can be used in many domains, the need to purchase expensive, specialized software and hardware to monitor the PoG limit the use of eye-tracking systems by consumers. In this paper, we describe an eye-tracking system that was integrated into a modern smartphone; the most widely accessible and deployed personal computing platform \[[@B27-sensors-20-00543]\] in history. The paper describes a smartphone-based eye-tracking system that is accurate and relatively insensitive to motion between the device and the user's head. Due to the increasing body of research regarding what can be learned about mental states of an individual though analysis of their gaze, high-quality eye-tracking becoming a pervasive sensor in most devices (just as GPS and accelerometers have) would present another layer of privacy concerns for users of these devices \[[@B28-sensors-20-00543]\]. However, the potential large positive impact of greater access to eye-tracking warrants continuing research and development of the core technology.
The most accurate video-based eye-tracking systems \[[@B29-sensors-20-00543]\] use methods that rely on artificial illumination of the eyes with infrared (IR) light and the estimation of eye features (e.g., the center of the pupil) in images that are captured by IR sensitive cameras, as shown in [Figure 1](#sensors-20-00543-f001){ref-type="fig"}. These systems can measure the PoG with bias that is lower than 0.5° \[[@B29-sensors-20-00543]\]. As IR illumination sources and IR sensitive cameras are not yet standard in commercial smartphones, all the prior work on smartphone-based eye-tracking system has used techniques that rely on visible light illumination of the eye (either by sun light or light emitted from the display) \[[@B30-sensors-20-00543],[@B31-sensors-20-00543],[@B32-sensors-20-00543],[@B33-sensors-20-00543],[@B34-sensors-20-00543],[@B35-sensors-20-00543],[@B36-sensors-20-00543],[@B37-sensors-20-00543]\]. A history of gaze enabled handheld mobile devices which use visible light illumination of the eye can be found in \[[@B28-sensors-20-00543]\]. Even under favorable operating conditions these systems have reported gaze bias that is larger than 3°. Smartphones may not always face the current limitations with respect to IR light sources and IR sensitive cameras, and indeed IR light sources and IR sensitive cameras have begun to appear in some commercial smartphones to enhance face tracking and authentication systems \[[@B38-sensors-20-00543],[@B39-sensors-20-00543]\]. While the manufacturers of these smartphones do not yet make the IR components available to third party software developers, the devices are otherwise essentially ready to enable IR-illuminated-based eye-tracking. Apple, as part of its augmented reality framework (ARKit \[[@B40-sensors-20-00543]\]) has now introduced calibration-free gaze estimation as an API for iPhoneX.
In this paper, we explore the performance and viability of accurate IR model-based eye-tracking in a smartphone the mobile form factor, which moves significantly relative to the user during use. We then compare those results to the best available mobile eye-tracking systems which are presented in the literature.
1.1. Smartphone-Based Eye-Tracking Systems That Use Visible Light {#sec1dot1-sensors-20-00543}
-----------------------------------------------------------------
Several eye-tracking systems that are based on a smartphone platform and use visible light are described in the literature \[[@B30-sensors-20-00543],[@B31-sensors-20-00543],[@B32-sensors-20-00543],[@B33-sensors-20-00543],[@B34-sensors-20-00543],[@B35-sensors-20-00543],[@B36-sensors-20-00543],[@B37-sensors-20-00543]\]. One of these systems, GazeCapture \[[@B35-sensors-20-00543]\], uses natural light (e.g., room illumination) and images that are captured by the front-face camera of the smartphone to generate estimates of the PoG. The system uses a deep convolutional neural network (CNN) to process images of the subject's face and the point-of-gaze estimate is produced by the CNN. In GazeCapture \[[@B35-sensors-20-00543]\] The CNN was trained with tens of thousands of labeled images from nearly 1500 people. The large training set that spans the expected range of natural variations in head pose and room lighting at home or in the office makes the system robust to novel faces in these environments. The GazeCapture \[[@B35-sensors-20-00543]\] eye-tracking system has a gaze-estimation bias that is greater than 3° \[[@B35-sensors-20-00543]\].
Another, smartphone-based eye-tracking system, ScreenGlint \[[@B36-sensors-20-00543]\], uses a corneal reflection that is created by the light emitted by the smartphone's display to generate PoG estimates that are based on the distance between the iris center and this corneal reflection. For a stationary eye-tracking system (i.e., the smartphone is not moving) ScreenGlint \[[@B36-sensors-20-00543]\] reported a gaze-estimation bias of 2.9°. As discussed in \[[@B29-sensors-20-00543]\] an approach that using a single corneal reflection and the center of the pupil/iris is not inherently robust to head or device motion, so the accuracy of the system will deteriorates during typical use when the smartphone will be handheld and free to move.
1.2. Smartphone-Based Eye-Tracking Systems That Use Artificial Infrared Illumination {#sec1dot2-sensors-20-00543}
------------------------------------------------------------------------------------
Gaze-estimation systems that require artificial infrared illumination and infrared cameras have demonstrated both high accuracy and robustness to motion \[[@B41-sensors-20-00543]\]. In these systems the location of eye features (the center of the pupil and two or more corneal reflections) are used to determine gaze position. These systems use a 3D gaze-estimation model \[[@B29-sensors-20-00543],[@B42-sensors-20-00543]\] which make the PoG estimates robust to relative motions between the head and the smartphone. In a previous study \[[@B43-sensors-20-00543]\] we used this approach to achieve an average gaze-estimation bias for a handheld mobile eye-tracker on a smartphone of 1°. Closer examination of our results showed that even though the gaze-estimation method is insensitive to movements between the eye-tracker and the subject's head, the bias in gaze estimations did change as a function of the distance between the head and the smartphone from between 0.4° to 2.1°. The main reason for these significant changes was the system's inability to accurately estimate the positions of the pupil center and corneal reflections at the full range of the expected relative positions of the head and the smartphone during regular use. Dramatic changes in the appearance of the face and the eyes during regular use make this task particularly challenging. In this paper, we address this issue by developing machine-learning algorithms (convolutional neural networks---CNNs) that can estimate the position of eye features accurately for the full range of expected smartphone motions. The approach is similar to that used for the detection and estimation of facial features \[[@B44-sensors-20-00543],[@B45-sensors-20-00543],[@B46-sensors-20-00543],[@B47-sensors-20-00543]\]. The eye-feature extractor uses a hierarchical cascade of CNNs and a center-of-mass output layer to estimate the position of the eye features. With this structure, the locations of the corneal reflections and pupil center can be estimated with sub-pixel accuracy and negligible bias while the computation is efficient enough to run on a smartphone. As our approach involves both the use of machine-learning models and geometric gaze-estimation models, we refer to this as a 'hybrid' approach to gaze estimation.
2. Materials And Methods {#sec2-sensors-20-00543}
========================
2.1. Overview of the End-to-End Mobile Eye-Tracking System {#sec2dot1-sensors-20-00543}
----------------------------------------------------------
The top-level overview of our hybrid infrared smartphone-based eye-tracking system, reviewed here, is similar to the first-generation system, described in \[[@B43-sensors-20-00543]\], and makes use of the gaze-estimation model presented in \[[@B29-sensors-20-00543]\]. The four principal software components of the eye-tracker are a head tracker, a feature extractor, a calibrator, and a gaze estimator. [Figure 2](#sensors-20-00543-f002){ref-type="fig"} illustrates the flow of data between the software modules. We run these software modules on an industrial infrared prototype smartphone, provided to us by Huawei \[[@B48-sensors-20-00543]\].
### 2.1.1. Infrared Smartphone {#sec2dot1dot1-sensors-20-00543}
The smartphone includes two infrared LEDs that illuminate the user's face together with and infrared-sensitive front-facing camera. Huawei \[[@B48-sensors-20-00543]\] integrated these components into an industrial prototype smartphone explicitly to explore mobile eye-tracking in future products. The front-facing camera has a 4 K resolution, which is similar to high-end current Android devices. It also has a 5-inch display, 1.2 GHz processor, and 1 GB of RAM, which makes it both smaller and significantly less powerful than modern phones. [Figure 3](#sensors-20-00543-f003){ref-type="fig"} illustrates the layout of our device.
### 2.1.2. Head Tracker {#sec2dot1dot2-sensors-20-00543}
A commercial head tracker \[[@B49-sensors-20-00543]\] is used to identify the presence of a face and the approximate locations of the eye regions. These regions initialize the feature extractor from a cold start or when local tracking fails. For this reason, the system uses the head tracker only occasionally. When the eye features were tracked successfully in the previous frame, the regions that initialize the feature extractor are centered on the estimation of the previous center of the pupil. Localization of narrow eye region windows is done though a hierarchy of small efficient CNNs as discussed in [Section 2.2](#sec2dot2-sensors-20-00543){ref-type="sec"}.
### 2.1.3. Calibrator {#sec2dot1dot3-sensors-20-00543}
The calibration procedure determines four physical geometric properties of each user's eyes as described in \[[@B29-sensors-20-00543]\]. These parameters are; (1) the radius of curvature of the cornea, (2) the distance between the center of corneal curvature and the center of the pupil, (3) the vertical offset angle between the optical and visual axis, and (4) the horizontal offset angle between the optical and visual axis. To perform calibration, and to estimate the values of these four parameters, a user is required to gaze at two (or more) known target locations. Then, starting with physiologically average human values as a seed, the four parameters are varied with a nonconvex optimizer. Gaze estimates are generated for each target and the average Euclidean distance between the gaze estimates and their corresponding targets is minimized. Once a calibration procedure has been completed once for a subject, these parameters can be saved and reused for future uses of the eye-tracker by the same person.
### 2.1.4. Gaze Estimator {#sec2dot1dot4-sensors-20-00543}
The gaze estimator uses a 3D gaze-estimation model that takes in the user-specific eye geometry \[[@B29-sensors-20-00543]\] and the locations of the pupil center and two corneal reflections in the eye-images to estimate the PoG. The gaze-estimation model is invariant to both device and head movements \[[@B29-sensors-20-00543],[@B42-sensors-20-00543]\].
### 2.1.5. Feature Extractor {#sec2dot1dot5-sensors-20-00543}
The feature extractor estimates the locations of the pupil center and two corneal reflections in eye-images from the front-facing camera of the smartphone. Its design is presented in the next section as it is a key contribution of this work.
2.2. Feature Extractor Architecture {#sec2dot2-sensors-20-00543}
-----------------------------------
The feature extractor consists of a hierarchy of multiple independently trained CNNs. One of the CNNs operates as an eye region classifier, and the rest directly estimate feature locations. The classifier network determines if both corneal reflections are visible and that the pupil is not significantly occluded (such as when a user blinks). If this network determines that the eye region is valid, several independent position estimation networks determine the location of the pupil center and corneal reflections.
The classifier and position estimation networks use similar base CNN architectures, as shown in [Figure 4](#sensors-20-00543-f004){ref-type="fig"}. They contain several convolutional layers (with batch normalization and optional pooling) followed by fully connected dense layers. Within this base architecture, there are many configurable hyperparameters (see [Table 1](#sensors-20-00543-t001){ref-type="table"}) but the primary architectural differences between the networks explored in this work occur at the output layer, the loss function, and the encoding of ground truth. The subsequent discussion of both the classifier and position estimation networks will focus on these.
### 2.2.1. Classifier {#sec2dot2dot1-sensors-20-00543}
The classifier is responsible for determining if an image of an eye region is valid (i.e., it contains two corneal reflections and a pupil). [Figure 5](#sensors-20-00543-f005){ref-type="fig"} illustrates examples of valid and invalid eye region images. The output layer for the classifier, shown in [Figure 6](#sensors-20-00543-f006){ref-type="fig"}, extends the base architecture with a dense layer containing two neurons, one for each of the two classes: invalid or valid. The final layer of neurons is followed by a SoftMax function to generate the probability that the given image belongs to each class. The ground truth for each training sample is a one-hot encoded vector indicating the class of the example as either a valid eye region or an invalid region. The loss function is a cross-entropy function that sums the log of the difference between each predicted class probability and the ground truth. This approach is typical for image classification \[[@B50-sensors-20-00543]\]. The cross-entropy equation for a binary classification is shown in Equation ([1](#FD1-sensors-20-00543){ref-type="disp-formula"}), where y is a binary indicator of valid/invalid eye region class label, p is the predicted probability of an eye region being valid, and N is the size of the training batch. No further processing happens for eye regions classified as invalid. Otherwise, the next step is to determine the locations of the pupil center and the two corneal reflections. $$\sum\limits_{i = 1}^{N} - \left( y_{i} \ast log\left( p_{i} \right) + \left( 1 - y_{i} \right) \ast log\left( 1 - p_{i} \right) \right)$$
### 2.2.2. Feature Estimator {#sec2dot2dot2-sensors-20-00543}
The position estimation networks are crucial for achieving high accuracy and robust mobile eye-tracking. A standard approach to position estimation is the construction of regression type networks, as illustrated in [Figure 7](#sensors-20-00543-f007){ref-type="fig"}. This approach extends the base architecture with a final two-neuron dense layer representing a predicted (x,y) relative location of the feature in the input image. For example, an inference output of \[0.5, 0.5\] would indicate the feature was in the center of the input image. The ground truth for each training sample is the human-labeled relative (x,y) location of the feature. The objective when training the network is to minimize the sum of the Euclidean distances between the predicted locations and the ground truths for a given training batch. It is computed with Equation ([2](#FD2-sensors-20-00543){ref-type="disp-formula"}), where N is the number of eye regions in the training batch, $x_{i}$ and $y_{i}$ are the estimated x and y location for a particular sample and $gtx_{i}$ and $gty_{i}$ are the ground truth labeled x and y location for a sample. At a higher level, we care about selecting a network which optimizes the accuracy of the position determination while being constrained by the computational and memory footprint limitations imposed by the desire to operate in real time on a smartphone. Two factors that play an important role here are the design of the output layer and the choice of the input image size. $$\sum\limits_{i = 1}^{N}\sqrt{\left( x_{i} - gtx_{i} \right)^{2} + \left( y_{i} - gty_{i} \right)^{2}}$$
The input image size plays an essential role in attaining both high accuracy and computationally efficient position estimation networks. It is vital to avoid excessive down-sampling the input image to achieve the highest accuracy because of the reduction of precision that such down-sampling implies. A direct, unscaled, crop of the original image near the eye region should be used if possible. [Figure 8](#sensors-20-00543-f008){ref-type="fig"} shows four examples of crop sizes that could be used for training a pupil center estimator. Crop size refers to the size of the window, which is taken from the original image, while down-scaling refers to how much that cropped image is reduced in size, through scaling, before sending it to the network. For example, if we compare a 256 × 256 crop down-scaled by four times with a native 64 × 64 crop that is not down-scaled (of the same eye region) both images have the same number of pixels. However, the size of the pupil in the native 64 × 64 crop will be four times bigger than the alternative. To maintain the same spatial resolution of the pupil we might choose not to downscale the 256 × 256 crop but now the network will be more expensive to run.
We explored four image crop sizes and determined that a smaller input image size (a tighter crop around the eye region) results in less overall computation and produces (as shown later in a [Section 3](#sec3-sensors-20-00543){ref-type="sec"}) a *more accurate* estimate of the pupil center. The tighter crop requires that the system must already know the approximate pupil center. To solve this problem, we used a sequence of small CNNs to focus in on the rough pupil-center position successively. We train these networks on aggressively sub-sampled (scaled) images of larger regions around the eye. In these networks, we trade accuracy for speed/computational effort, because we only need to determine an area that contains the eye and not the precise location. Using this type of hierarchy, we estimate the pupil center to within a few pixels. Then, from a very tightly cropped region around that location, more computationally expensive and precise locator networks find the exact pupil center and corneal reflections. A key feature of which is the construction of the output layer.
The output layer of a regression network can also have a positive impact on performance, which is to say variations here can improve the position estimation accuracy while only negligibly increasing overall computation or memory footprint of the entire network (including the base architecture). In this work, the output layer is a dense layer with one neuron per pixel of the input image, followed by a SoftMax layer, and a two-dimensional center-of-mass computation ([Figure 7](#sensors-20-00543-f007){ref-type="fig"}). In this approach, we treat the position estimation problem as a classification problem by making a network which maps an output neuron to each input pixel. Each neuron outputs the probability that any specific pixel was the target location. During a forward inference pass, there are many pixels in the area around ground truth location that will have some significant non-zero probability of being the inferred feature location. Therefore, rather than merely selecting the pixel with the highest probability, we apply a two-dimensional center-of-mass calculation to that set of probabilities.
In [Section 3](#sec3-sensors-20-00543){ref-type="sec"}, we will show that our center-of-mass output architecture improves feature estimation accuracy to computational cost ratio relative to the naive regression approach. [Figure 9](#sensors-20-00543-f009){ref-type="fig"} shows the complete architecture of our feature extractor hierarchy. It consists of six total networks: two very fast, but approximate, coarse pupil estimators which localize successively tighter eye regions. Next is a classifier to determine if the generated tight crop is a valid eye region. Finally, the classifier is followed by three accurate position estimators to determine the exact position of the pupil center and both corneal reflections. These positions are then given to the gaze-estimation model to compute the PoG.
The next section discusses how we trained, evaluated, and selected the specific feature extractor networks that we eventually integrated into our end-to-end eye-tracking system.
3. Feature Extractor Training And Performance {#sec3-sensors-20-00543}
=============================================
In this section, we outline the training, analysis, and selection of the feature extractor CNNs described previously, beginning with the collection of a dataset for training.
3.1. Dataset Collection {#sec3dot1-sensors-20-00543}
-----------------------
Our training dataset consists of infrared images of eye regions from the smartphone's front-facing camera with labeled positions of the eye features. A dataset of this kind did not yet publicly exist, so we had to collect a new one. The image variability we expect in a smartphone environment comes mainly from the relative position and orientation changes between the device and the subject's face. There can also be higher variability because of the (possibly) weaker contrast between the pupil and iris, which is caused by limited illumination power and a noisier camera sensor.
We collected and labeled a total of 2000 face images (4000 eye regions) from 100 participants without glasses (without controlling for contacts) holding our prototype infrared smartphone. Each participant was asked to position the device anywhere that they felt comfortable and look at a random target on the display. The device captured a single photo. Then the subject was given five seconds to re-position the device and repeat the same procedure before the next image was captured. The positioning and re-positioning of the device resulted in a variety of relative distance and orientations of the device and the participants' heads. This process repeated 20 times for each subject, resulting in 20 infrared images of their face captured by the front-facing camera. During the 20-image collection process a random set of 6 frame indexes are generated for the purpose of artificially increasing the number of invalid eye regions in the dataset. When capturing these specific frames either one or both infrared LEDs (chosen at random) are disabled resulting in eye regions unsuitable for the gaze-estimation model used in this work. For the other 14 frames both LEDs were enabled. The orchestration of the LEDs in this fashion was done automatically by the sample collection app.
Re-positioning of the device resulted in a good variation of the relative distance and orientation between the device and the participants' head. If the subject held the device in such a way that their eyes were not somewhere in the frame than they were informed of this by the experimenter In practice this occurred less than 10 times in the collection of 2000 samples. This is due, in part, to the wide field of view on the front-facing camera of our prototype smartphone of 75° combined with the active task of looking at a target. Data collection occurred exclusively in indoor environments with minimal natural light interference (although no specific controls were employed to avoid occasional direct sunlight from windows).
In the first stage of labeling each eye region is labeled with either a valid eye or an invalid eye class label. An eye region is considered 'invalid' if the pupil is significantly occluded, a corneal reflection is missing, or the image is exhibiting significant motion blur. Invalid eye regions can naturally occur during a blink or when images are captured during large device motion. Examples of eye regions labeled as valid and invalid are shown in [Figure 5](#sensors-20-00543-f005){ref-type="fig"}. The class distribution between valid and invalid eye regions in the 2000 captured images is approximately balanced, with 55% valid and 45% invalid. In many cases, eye regions are either clearly valid or clearly invalid by the criteria mentioned above. Qualitatively approximately 1--2% of eye regions were difficult for the human annotator to classify. These cases require a judgement as to what is too much occlusion or too much motion blur. There is not a right or wrong answer here, but it is important to highlight the existence of such cases and their approximate frequency. The upper limit of the accuracy of any classification algorithm applied to this dataset will be limited by this frequency.
In the second stage of labeling, the locations of the pupil center and both corneal reflections were determined manually for each of the valid eye regions. As corneal reflections are small and well defined in the images they were labeled by the annotator with a single click. Single click estimation introduces quantization errors associated with the mapping of a continuously varying corneal reflection center location with an integer \[x,y\] pixel location. The labeling procedure will result in a quantization error of approximately 0.25 pixels in both the x and y direction, or an average error magnitude of approximately 0.35 pixels (assuming the probability density function of the quantization error is uniform). To label the pupil center the annotator first clicks on 10 points along the pupil--iris boundary. These 10 boundary points are used by the OpenCV \[[@B51-sensors-20-00543]\] computer vision and image processing library function cvfitellipse2 to fit an ellipse to the pupil--iris boundary. The center of the fitted ellipse is recorded as the ground truth pupil center location for our dataset. The uncertainty in the pupil location estimates was determined empirically by selecting an eye window from ten different subjects and labeling each of them ten separate times. The average magnitude of the deviation from the mean of each pupil estimate was 0.18 pixels horizontally and 0.29 pixels vertically (or 0.34 pixels total magnitude). Increased uncertainty in the vertical direction is expected due to interference with the upper eyelid occluding part of the pupil. Overall however, the labeling of the center of the pupil--iris boundary has sub-pixel accuracy.
From these 4000 labeled eye regions, we created four distinct datasets, each with one of the following image crop sizes around a random location near the center of the labeled pupil center: 64 × 64, 96 × 96, 128 × 128, and 256 × 256 pixels. These data sets allowed us to evaluate the performance of the feature extractor networks with different crop sizes around the eye. Finally, we used data augmentation techniques to artificially increase the size of each dataset from 4000 eye regions to 40,000 eye regions by including ten crops at ten random locations near each labeled pupil center rather than only one. We split each dataset into separate training and validation sets which contain eye regions from 80 and 20 of all the subjects (100), respectively. Testing the networks was done by using data that were generated during the experimental testing of the eye-tracker (i.e., the data were not used for training), which will be discussed in [Section 4](#sec4-sensors-20-00543){ref-type="sec"}.
3.2. Classifier Network Performance {#sec3dot2-sensors-20-00543}
-----------------------------------
The classifier is responsible for determining if an image of an eye region is valid (i.e., it contains two corneal reflections and a pupil). To evaluate and select the parameters for the classifier network, we performed the following experiment. First, we generated 256 hyperparameter configurations of the base architecture from the set shown in [Table 1](#sensors-20-00543-t001){ref-type="table"}. Any configurations which required more than the available 11 GB of GPU memory to initialize and train (for a 256 × 256 image crop with a batch size of 32) were discarded and new configurations generated to replace them. The range of parameters in [Table 1](#sensors-20-00543-t001){ref-type="table"} were selected such that the inference networks generated would cover a wide range of computational complexities. The smallest possible network (one CNN layer with four 3 × 3 kernels, 4× downscale, and no hidden fully connected layers) can run on any modern smartphone at high number of frames per second (\>100). The largest possible network (five CNN layers each with 128 9 × 9 kernels followed by hidden fully connected layers of 2048 neurons each) is much larger than the largest networks in Googles MobileNetV2 \[[@B52-sensors-20-00543]\] family of image processing networks, which runs at between 3 and 30 ms \[[@B53-sensors-20-00543]\] per inference depending on the device and accelerator used. As we aimed to achieve an estimation rate of at least 30 frames/s cumulatively across all our networks and for both eyes (at least theoretically if our device had modern accelerators) our exploration of the trade-offs between inference quality and inference cost for a smartphone-based eye-tracker occurred between these two extremes.
Then we trained each specific architecture independently with the four datasets discussed previously. Each network was trained with a batch size of 32 for a total of 20 epochs. To permit early stopping and minimize overtraining, at the end of each epoch the network was saved only if the absolute difference between the average training and validation errors for that epoch was smaller than after all previous epochs. Each network was evaluated based on two metrics; the classification accuracy on the validation set, and the computational cost of one forward inference though the network. We measured this cost in millions of floating-point operations (mflops), and this value was provided automatically by the neural network training library that we used Tensorflow version 1.12 \[[@B54-sensors-20-00543]\].
[Figure 10](#sensors-20-00543-f010){ref-type="fig"} shows the Pareto-optimal networks for this experiment, where each point in the figure is a specific instance of a trained network. The *x*-axis is the classification accuracy using a probability threshold of 0.5, thereby for an eye region to be classified as "valid" the output probability of the network for that class was over 50%. The four curves correspond to the size of the input image crop that was used to train and validate data for each curve. Observe that the bottom curve, corresponding to a 64 × 64 input image crop, is superior to all the other curves. The implication is that training on a smaller eye region crop produces similar or better classification accuracy with much less computational effort. As an example, to achieve a classification accuracy of 98% on the 128 × 128 pixel dataset required a network that used roughly 30 times more computation than to achieve the same accuracy on the 64 × 64 dataset.
As our manual labeling of the images requires a judgement as to when an image has too much pupil occlusion the final chosen configuration, which achieved accuracy of 98%, is near perfect given the slight subjectivity of the human-annotated classification labeling.
3.3. Feature Extractor Network Performance {#sec3dot3-sensors-20-00543}
------------------------------------------
We evaluated the pupil center and corneal reflections position estimation networks similarly to the classifier above. We selected 256 hyperparameter configurations for the base architecture randomly from the set of choices outlines in [Table 1](#sensors-20-00543-t001){ref-type="table"}. Each network was trained with a batch size of 32 for a total of 20 epochs. To permit early stopping and minimize overtraining, at the end of each epoch the network was saved only if the absolute difference between the average training and validation errors for that epoch was smaller than after all previous epochs. Each network was evaluated based on the mean distance (measured in pixels of the original image) between the networks output estimate and the ground truth locations in the validation set. For brevity, this is shown primarily for the pupil-center estimation, although results are similar for the corneal reflections.
[Figure 11](#sensors-20-00543-f011){ref-type="fig"} shows the Pareto-optimal curves for the suite of pupil estimation networks using the center-of-mass output layer described in [Section 2.2](#sec2dot2-sensors-20-00543){ref-type="sec"}. The *x*-axis is the mean estimation error (in pixels), and the *y*-axis is the number of mflops required to compute a single estimate. We reported the mean pupil estimation errors in this figure after eliminating estimates that had significant errors greater than 7.5 pixels. The removal of large errors is done to improve the comparison between networks trained on different image crop sizes. Otherwise, this type of analysis would favor smaller image crop sizes because gross errors in larger image crops are higher (in the absolute number of pixels) than gross errors in smaller image crops. Even with this correction, [Figure 11](#sensors-20-00543-f011){ref-type="fig"} shows significant advantages for networks trained with smaller crop sizes. These advantages extend to both the performance-to-cost ratio and absolute estimation accuracy. This result led us to select the 64 × 64 image crop size for our pupil-center estimation network.
The improvement in both the computational load *and* the mean estimation error when training networks with tighter crop sizes is an interesting and counter-intuitive result. As a 64 × 64 crop is a direct subset of a 256 × 256 crop (of the same eye) in principle networks trained on the larger crop should be slower but get equivalent results. However, the 256 × 256 crop has extra information which is not relevant to the estimation of the pupil--iris boundary (the only feature relevant to estimating the pupil center) and thus the larger crop size presents an image with more nuisance parameters (i.e., lower signal-to-noise ratio). Training length, learning rate and optimizer explorations did not eliminate this effect. Training with 10--100 times more data might have an impact on this effect, but that effort is beyond the scope of this paper. In practice, when training small networks with small datasets it is important to limit the number of nuisance parameters."
The next experiment was to evaluate the performance of the pupil position estimator with and without the center-of-mass output layer. [Figure 12](#sensors-20-00543-f012){ref-type="fig"} shows the Pareto-optimal curves for networks using simple regression and networks using the center-of-mass output layer. Across all 256 random hyperparameter configurations for the 64 × 64 cropped dataset, the average estimation error was improved by 14% by using the center-of-mass calculation. Near the best accuracy to cost ratios of the Pareto-optimal curves, this improvement was roughly 10%. Combining the improvements from both tighter image crops and the center-of-mass output layer, the net reduction in pupil estimation error is about 50% (at any given amount of mflops) when compared with a naive single CNN regression network trained on a 256 × 256 image crop.
Finally, we selected one specific configuration for each of the six networks in the feature extractor. Recall that this system consists of two small locator networks to quickly localize the pupil, a classifier to validate that the cropped image included a "valid" eye, and three accurate position estimators to estimate the pupil center and corneal reflections precisely. We selected these specific networks from points on the corresponding Pareto-optimal curves which represent the good trade-off between performance (error) and computational cost. [Table 2](#sensors-20-00543-t002){ref-type="table"} shows the image crop sizes, amount of down-scaling where applicable, average absolute error (in native pixels of the original unscaled image crop) between predictions and feature labels on our validation set, and computational cost of these networks (in megaflops).
For the course pupil and medium pupil networks (used to localize the 64 × 64 eye region crop) we explored down-scaling factors up to a maximum of eight rather than four because those networks required less spatial accuracy and needed to be as fast as possible. This localization step could have been done with a single 256 × 256 crop size network, but would have been more expensive to achieve the same average pupil estimation error. In our dataset the size of bounding boxes centered at the pupil center and containing both the entire pupil and both corneal reflections were as large as 58 pixels (for subjects with large pupils holding the phone close to their face). As a result, we would like most of the pupil localization estimates to have less than 6 pixels of error horizontally and vertically to avoid clipping part of the required eye features. This was more cost effective to achieve with a two-stage hierarchy.
There was significant flexibility in the selection of the hyperparameters that generated the best results. In most cases, we could have selected a different network on the Pareto curves that looked quite different but performed very similarly. For example, a network with similar performance might have had more CNN layers, but fewer kernels per layer, resulting in approximately the same overall computational requirements and accuracy. With that said, [Table 3](#sensors-20-00543-t003){ref-type="table"} shows the specific architecture parameters for the networks we selected for completeness. For each of the five networks [Table 3](#sensors-20-00543-t003){ref-type="table"} presents the number of kernels (k) and kernel width (w) for each CNN layer, the number of neurons in the hidden dense layer, and the dropout percentage between the hidden dense layer and the output layer. Pooling was enabled for all the networks we selected. The positive effect of pooling is not surprising for the classifier networks but was surprising for the feature estimation networks. It seems that the loss in spatial resolution from pooling CNN output feature maps was less critical than the gains from adding more kernels when normalized for total compute requirements. As a result, all selected networks had pooling enabled for every CNN layer. In the next section, we use the above networks to evaluate the end-to-end mobile eye-tracker.
4. Eye-Tracking Results and Comparisons {#sec4-sensors-20-00543}
=======================================
In this section, the performance of the hybrid eye-tracking system is evaluated and compared with three other mobile systems: Screen Glint \[[@B36-sensors-20-00543]\], GazeCapture \[[@B35-sensors-20-00543]\], and the previous version of our system \[[@B43-sensors-20-00543]\]. We present these comparisons through a series of experiments. The 8 subjects who participated in both experiments did not contribute data to either the training or validation section of the dataset, which was described in [Section 3.1](#sec3dot1-sensors-20-00543){ref-type="sec"}.
Gaze-estimation models for stationary eye-tracking systems that are similar to the one that we used for our mobile eye-tracker have been extensively evaluated in the literature \[[@B29-sensors-20-00543],[@B41-sensors-20-00543],[@B55-sensors-20-00543],[@B56-sensors-20-00543]\] with a variety of camera and light source configurations. These studies demonstrated that when robust eye features are provided to this 3D gaze-estimation model, high-quality gaze estimates can be produced in the presence of relative motion between the subject and the system. The purpose of the experiments in this section are to validate how well inexpensive CNNs can produce robust eye features in the presence of larger relative motion than was exhibited in these previous studies and when willing participants use a smartphone naturally in an indoor environment for the purpose to eye-tracking. As a result, the number of participants in this part of the study did not need to be large.
4.1. Eye-Tracker Performance at Different Distances {#sec4dot1-sensors-20-00543}
---------------------------------------------------
During the evaluation of the eye-tracker performance subjects were instructed to hold the smartphone in any position or orientation they felt comfortable. Then, the experimenter adjusted only the distance between the eye-tracker and their head to be approximately 30 cm at which point a calibration procedure was performed. During this calibration procedure five targets were shown on the display, one at a time, in the 'plus' pattern as shown in [Figure 13](#sensors-20-00543-f013){ref-type="fig"}a. As the subject gazed at each target the system collected 50 sets of eye features. The Levenberg Marquardt nonconvex optimization algorithm \[[@B57-sensors-20-00543]\] was used to estimate the subject specific parameters which minimized the distance between the gaze estimates generated with those eye features and the corresponding target locations. These subject specific parameters were saved and used in the next part of the experiment.
Once the calibration parameters were estimated the experimenter adjusted distance between the eye-tracker and the subjects' head to one of five approximately fixed distances; 20, 25, 30, 35, and 40 cm. Subjects' were instructed to avoid large intentional movements, but that small natural head and device movement was both acceptable and expected. At each of the five distances each subject was asked to look at five target locations. The targets were displayed one at a time in the pattern shown in [Figure 13](#sensors-20-00543-f013){ref-type="fig"}b. For each target, the system collected 50 eye-gaze estimates using the calibration parameters generated in the previous step.
The average distance (in mm) between a target location and the mean location of 50 gaze estimates on each target were used as an estimate of a gaze bias. This value was converted into degrees using the approximately known distance between the subject and the device and then averaged across all 5 targets. [Table 4](#sensors-20-00543-t004){ref-type="table"} presents the average gaze-estimation bias, in degrees, at each distance and compares that to other mobile eye trackers. In that table, we refer to our previous work \[[@B43-sensors-20-00543]\] as SmartEye and the current work as HybridTrack. GazeCapture did not present results at fixed distances, so for comparison, we only provided the average gaze bias. To more fairly compare this work with our previous work we reran SmartEye to produce gaze estimates on the same videos used in this work. As a result, the gaze bias results for SmartEye in [Table 4](#sensors-20-00543-t004){ref-type="table"} are slightly different than what was presented for a similar experiment in \[[@B43-sensors-20-00543]\], but show the same trends.
It is clear from [Table 4](#sensors-20-00543-t004){ref-type="table"} that both the SmartEye and the HybridTrack have much lower gaze estimation bias than ScreenGlint or GazeCapture. The comparison between SmartEye and the present work also shows the improvement in robustness of the new mobile eye-tracker as it can maintain high gaze-estimation accuracy in the presence of significant distance variation. Please note that at the calibration distance of 30 cm, SmartEye performed very well (better than HybridTrack) with a gaze bias of only 0.57${}^{\circ}$ gaze. However, changing the distance by 10cm resulted in significant performance degradation (bias to as high as 1.90${}^{\circ}$). The new HybridTrack eye-tracker produced gaze-estimation bias results with much smaller variation (0.65${}^{\circ}$ to 0.81${}^{\circ}$) over the range of distances. An individual break down of the gaze performance for each of the 8 subjects tested in this work is shown in [Table 5](#sensors-20-00543-t005){ref-type="table"}. This table also includes the eye color of each subject (light for blue hues, dark for brown hues). The contrast between the pupil and iris is worse for users with blue irises since their irises do not reflect as much IR as subjects with brown irises. This can lead to less robust feature extraction of the pupil--iris boundary and less accurate gaze estimates. The use of a CNN feature extractor trained on a wide range of iris colors mitigated this effect as the final gaze-estimation errors are similar for both groups. The sample size of only two light eye subjects would need to be increased however to come to any definitive conclusions.
4.2. Unsupervised Eye-Tracker Use {#sec4dot2-sensors-20-00543}
---------------------------------
The second experiment was designed to evaluate the performance of the hybrid eye-tracker from the perspective of a software developer who wants to make use of the tracker in a broader application. When deploying an application that uses eye-tracking, the user can be given guidance on to how to hold and use the device, but its use would be in an unsupervised environment where the position and orientation of the device are unknown. Additionally, when the eye-tracking system indicates that a subject is looking at a specific target, the application developer would be interested in the certainty of such an observation.
With this in mind, we designed the following experiment. We segmented the screen of the display into a grid of boxes that are approximately square target regions. Then, in a random order, each grid box (target) was highlighted, indicating to the subjects that they should direct their gaze to that region. Following a 500 ms delay, the system recorded 15 gaze estimates. We refer to this procedure as a grid test. One metric derived from this test is the percentage of gaze estimates that were inside the highlighted box. It is common, in eye-tracking, to also compute a single gaze estimate from a sequence of *N* consecutive estimates, to reduce RMS error through averaging. We show the results of averaging 2 to 10 estimates in [Figure 14](#sensors-20-00543-f014){ref-type="fig"}.
The specific experimental procedure for the grid test was as follows. Each one of eight subjects who participated in the study was asked to hold the device comfortably. Next, the subject performed a calibration procedure. The calibration parameters were stored and used in each subsequent session with that subject. Then, at three different times over the next five days, each subject was asked to perform the grid test at four different grid sizes; 3 × 2 (38.3 × 32.5 mm boxes), 4 × 3 (28.8 × 21.6 mm boxes), 6 × 4 (19.2 × 16.3 mm boxes), and 8 × 6 (14.4 × 10.8 mm boxes). After being given the device and the instructions, the test involved no experimenter supervision.
The calibration procedure of the infrared 3D gaze-estimation model used in this work estimates physiological geometries of a subject's eye, which does not change over time. Therefore, we do not need to re-calibrate between each use of the eye-tracker after large relative movements between the subject and the device. This is a key property of our system when compared with other mobile eye trackers.
[Figure 14](#sensors-20-00543-f014){ref-type="fig"} shows the percentage of gaze estimates that were within the correct target boxes as a function of the number of averaged estimates. Each point in [Figure 14](#sensors-20-00543-f014){ref-type="fig"} is the average data for all 8 subjects for each experiment (i.e., a specific box size and number of averaged estimates). Take, for example, the 6 × 4 grid-size with number of averaged estimates of 4, [Figure 14](#sensors-20-00543-f014){ref-type="fig"} indicates a 90% estimation accuracy. Given that we collected 15 gaze estimates for each target box, there are 11 gaze estimates for each target that use an average of 4 estimates. An accuracy of 90% indicates that of the 2112 total gaze estimates that use an average of 4 gaze estimates (11 estimates per target × 24 targets per subject × 8 subjects) about 1900 of them are located in the correct corresponding target boxes. Each point on the graph represents the mean performance of the eight participants. The error bars in [Figure 14](#sensors-20-00543-f014){ref-type="fig"} show the standard error of the means
For the 3 × 2 grid (38.3 × 32.5 mm) [Figure 14](#sensors-20-00543-f014){ref-type="fig"} indicates that 95--100% of all gaze estimates (depending on the number of gaze estimates that were averaged) were inside the highlighted boxes. The percentage drops with each increase in grid dimension and the corresponding decrease in box size: 89--96% for the 4 × 3 grid (28.8 × 21.6 mm boxes), 83--93% for the 6 × 4 grid (19.2 × 16.3 mm boxes) and 66--76% for the 8 × 6 grid (14.4 × 10.8 mm boxes). The performance difference between the 6 × 4 and 8 × 6 grid sizes indicates the limit of the effective performance of our system and testing smaller grid sizes would not be valuable.
[Figure 15](#sensors-20-00543-f015){ref-type="fig"} shows the results of the grid test on an 8 × 6 grid (14.4 × 10.8 mm boxes) by giving the percent correct estimations at each grid location. These results are across all 8 subjects with N (the number of gaze estimates that are averaged) set to 10. While the average across all boxes was 76%, one can observe significant variations in performance for specific locations on the screen. Gaze estimates in the central area of the display are more likely to be correct, with percentages in the high 70 s to low 80 s. Estimates closer to the corners of the display can be quite less, with the worst occurring at the bottom left corner box, with only 38% being correct within the box. We suggest that future authors also present this level of spatial gaze-estimation detail in their evaluation of mobile eye trackers, rather than a single average estimation bias.
We hypothesize that the principal explanation for this effect is associated with the physical locations of the LEDs on the prototype smartphone. The two LEDs on our prototype are near the top right and bottom left of the display, and these are the two regions with the worst performance. When a subject gazes near these regions one of the corneal reflections, being furthest from the pupil center, may not be created by the spherical region of the cornea. Our model \[[@B29-sensors-20-00543]\] assumes that the corneal reflection is created by a spherical cornea, and so this can result in additional gaze bias. One solution to this problem on desktop systems is to have more LEDs (more than two) and to select the two corneal reflections closest to the pupil center when estimating the gaze. This solution might be less desirable on a smartphone due to power and space constraints.
5. Conclusions and Future Work {#sec5-sensors-20-00543}
==============================
We have presented a new hybrid eye-tracking system for smartphone. The new system can achieve gaze-estimation bias of 0.72${}^{\circ}$ in realistic unsupervised scenarios across 8 subjects. This performance is achieved with a single calibration, operating over a realistic range of device positions (20--40 cm) and orientations. The hybrid eye-tracking system uses machine-learning algorithms to improve the robustness and accuracy of eye-feature estimation (the location of corneal reflections and the center of the pupil). The eye features are then used by a gaze-estimation model that is insensitive to relative motion between the subject and the smartphone. When handing the device to subjects with little to no instructions, the system could achieve better than 90% accuracy distinguishing between targets in a 6 × 4 grid (using simple averaging of five or more gaze estimates). Our hybrid eye-tracking system (which has the advantage of infrared illumination) achieves significantly better accuracy (more than 400% better) than the accuracy achieved by previous systems that use natural light and appearance-based methods to estimate gaze position. The improvement in performance comes from the hybrid approach: a machine-learning feature extraction step followed by a geometric 3D model. This improvement may enable mobile applications that require analysis of visual scanning behavior of subjects that view a limited number of items on a smartphone screen.
An infrared mobile eye-tracking system has several limitations, however, and its performance will be affected by operational parameters not explored in this work. To achieve the performance reported in this paper the user must be holding the device such that the head/eyes are in the frame captured by the front-facing camera of the handheld device and neither are occluded by clothing or accessories \[[@B28-sensors-20-00543]\]. Infrared eye-tracking will naturally suffer from direct sunlight interference and is generally not suitable in bright outdoor environments when operated in direct sunlight. Operational parameters such as rapid changes in eye illumination due to rapid changes in the distance between the smartphone and the subject, shakiness resulting from use during transport, and reflections of the IR light from eye-glasses would all likely cause a reduction in performance. Solutions to these issues can be both use-case and hardware dependent. One possible general approach to help overcome these limitations however is to augment our infrared-illumination-based hybrid approach with gaze-estimation techniques that use natural light \[[@B35-sensors-20-00543],[@B36-sensors-20-00543],[@B58-sensors-20-00543]\]. In a dual-model system, situations when infrared eye features are difficult to localize the less accurate methods that use natural light can be used to improve the range of mobile eye-tracking-based applications. Another limitation of our eye-tracking system is the need to have a smartphone with a front-facing infrared camera and at least two infrared LEDs. This configuration is currently available only on a limited number of high-end devices, and the access to this hardware is only available to the device manufacturers.
Both a prototype infrared smartphone and a research grant which enabled this work was provided by Huawei Technologies Co., Ltd.
Funding acquisition, J.R. and M.E.; Investigation, B.B.; Methodology, B.B.; Software, B.B.; Supervision, J.R. and M.E.; Writing---original draft, B.B.; Writing---review & editing, J.R. and M.E. All authors have read and agreed to the published version of the manuscript.
This research was funded by an NSERC Collaborative Research and Development Project (CRDPJ) with Huawei Technologies Canada, and NSERC grant 480479.
The authors declare no conflict of interest.
All subjects gave their informed consent for inclusion before they participated in the study.
{#sensors-20-00543-f001}
![Eye-Tracking System \[[@B42-sensors-20-00543]\], licensed under <https://creativecommons.org/licenses/by/4.0/>.](sensors-20-00543-g002){#sensors-20-00543-f002}
{#sensors-20-00543-f003}
{#sensors-20-00543-f004}
{#sensors-20-00543-f005}
{#sensors-20-00543-f006}
{#sensors-20-00543-f007}
{#sensors-20-00543-f008}
{#sensors-20-00543-f009}
{#sensors-20-00543-f010}
{#sensors-20-00543-f011}
{#sensors-20-00543-f012}
![Sample Target Patterns Used for Calibration (**a**) or Estimation (**b**) (Adapted from \[[@B42-sensors-20-00543]\]).](sensors-20-00543-g013){#sensors-20-00543-f013}
{#sensors-20-00543-f014}
{#sensors-20-00543-f015}
sensors-20-00543-t001_Table 1
######
Base Architecture Hyperparameter Ranges.
Hyperparameter: Choice Set
------------------------------------ -----------------------------------
Input Down-Scaling Factor: \[1×, 2×, 4×\]
Number of CNN Layers: \[1, 2, 3, 4, 5\]
Number of Kernels in CNN Layer: \[4, 8, 16, 32, 64, 128\]
Width of Kernels in CNN Layer: \[3, 4, 5, 6, 7, 8, 9\]
Pooling at Each CNN Layer (2 × 2): \[enabled, disabled\]
Number of Hidden Dense Layers: \[0, 1, 2\]
Number of Neurons in Dense Layer: \[64, 128, 256, 512, 1024, 2048\]
Dropout % of Dense Layers: \[0.3, 0.4, 0.5, 0.6, 0.7\]
sensors-20-00543-t002_Table 2
######
Selected Networks: Size, Performance, and Cost.
-----------------------------------------------------------------------
Features Estimators Crop Size\ Downscale Error\ Cost\
(px) (native px) (mflops)
--------------------- ------------ ----------- ------------- ----------
pupil coarse 256 8× 3.818 2.5
pupil medium 128 4× 1.602 2.51
pupil fine 64 n/a 0.746 31.95
cr-a 64 2× 0.515 11.64
cr-b 64 2× 0.529 12.55
accuracy
classifier 64 n/a 98% 20.07
-----------------------------------------------------------------------
sensors-20-00543-t003_Table 3
######
Selected Networks: Architectures.
Network cnn l1 cnn l2 cnn l3 cnn l4 Dense Cost (mflops)
--------- -------- -------- -------- -------- ------- --------------- ----- ------- ------ ----- -------
p1 24 4 × 4 2 3 × 3 16 5 × 5 n/a n/a 512 0.7 2.50
p2 24 4 × 4 2 3 × 3 16 5 × 5 n/a n/a 1024 0.7 2.51
p3 8 4 × 4 32 6 × 6 8 3 × 3 32 6 × 6 1024 0.7 31.95
cr-a 64 5 × 5 8 3 × 3 64 5 × 5 n/a n/a 1024 0.5 11.64
cr-b 16 4 × 4 48 3 × 3 48 6 × 6 n/a n/a 1024 0.4 12.55
sensors-20-00543-t004_Table 4
######
Varied Eye-Tracker Depth.
System 20 cm 25 cm 30 cm 35 cm 40 cm Average
------------- ------------------ ------------------ ------------------ ------------------ ------------------ ------------------
GazeCapture n/a n/a n/a n/a n/a 2.98${}^{\circ}$
ScreenGlint n/a 3.38${}^{\circ}$ 3.32${}^{\circ}$ n/a 2.32${}^{\circ}$ 3.01${}^{\circ}$
SmartEye 1.90${}^{\circ}$ 1.15${}^{\circ}$ 0.57${}^{\circ}$ 0.90${}^{\circ}$ 0.98${}^{\circ}$ 1.10${}^{\circ}$
HybridTrack 0.81${}^{\circ}$ 0.70${}^{\circ}$ 0.71${}^{\circ}$ 0.73${}^{\circ}$ 0.65${}^{\circ}$ 0.72${}^{\circ}$
sensors-20-00543-t005_Table 5
######
HybridTrack Subject Specific Gaze Bias at Varied Eye-Tracker Depth.
Subject Eye Color \[Dark/Light\] 20 cm 25 cm 30 cm 35 cm 40 cm Average
---------- -------------------------- ------------------ ------------------ ------------------ ------------------ ------------------ ------------------
01 dark 0.66${}^{\circ}$ 0.58${}^{\circ}$ 0.60${}^{\circ}$ 0.59${}^{\circ}$ 0.63${}^{\circ}$ 0.61${}^{\circ}$
02 light 0.79${}^{\circ}$ 0.81${}^{\circ}$ 0.74${}^{\circ}$ 0.66${}^{\circ}$ 0.57${}^{\circ}$ 0.71${}^{\circ}$
03 dark 0.85${}^{\circ}$ 0.69${}^{\circ}$ 0.62${}^{\circ}$ 0.66${}^{\circ}$ 0.51${}^{\circ}$ 0.66${}^{\circ}$
04 dark 0.88${}^{\circ}$ 0.75${}^{\circ}$ 0.81${}^{\circ}$ 0.83${}^{\circ}$ 0.64${}^{\circ}$ 0.78${}^{\circ}$
05 dark 0.73${}^{\circ}$ 0.70${}^{\circ}$ 0.72${}^{\circ}$ 0.80${}^{\circ}$ 0.62${}^{\circ}$ 0.71${}^{\circ}$
06 light 0.79${}^{\circ}$ 0.65${}^{\circ}$ 0.62${}^{\circ}$ 0.64${}^{\circ}$ 0.72${}^{\circ}$ 0.69${}^{\circ}$
07 dark 0.88${}^{\circ}$ 0.74${}^{\circ}$ 0.79${}^{\circ}$ 0.83${}^{\circ}$ 0.70${}^{\circ}$ 0.79${}^{\circ}$
08 dark 0.95${}^{\circ}$ 0.73${}^{\circ}$ 0.81${}^{\circ}$ 0.79${}^{\circ}$ 0.78${}^{\circ}$ 0.81${}^{\circ}$
combined n/a 0.81${}^{\circ}$ 0.70${}^{\circ}$ 0.71${}^{\circ}$ 0.73${}^{\circ}$ 0.65${}^{\circ}$ 0.72${}^{\circ}$
|
{
"pile_set_name": "PubMed Central"
}
|
Q:
query to give workflow statistics like source count,target count,start time and end time of each sessions
I have one workflow which contain five sessions. I am looking for a query by using informatica repository tables/views which give me output like below. I am not able to get a query which give me desired result.
workflow-names session-names source-count target-count session-start time session-end time.
A:
If you have access to Repository metadata tables, then you can use below query
Metadata Tables used in query:
OPB_SESS_TASK_LOG
OPB_TASK_INST_RUN
OPB_WFLOW_RUN
Here the Repository user is INFA_REP, and workflow name is wf_emp_load.
SELECT w.WORKFLOW_NAME,
t.INSTANCE_NAME,
s.SRC_SUCCESS_ROWS,
s.TARG_SUCCESS_ROWS,
t.START_TIME,
t.END_TIME
FROM INFA_REP.OPB_SESS_TASK_LOG s
INNER JOIN INFA_REP.OPB_TASK_INST_RUN t
ON s.INSTANCE_ID=t.INSTANCE_ID
AND s.WORKFLOW_RUN_ID=t.WORKFLOW_RUN_ID
INNER JOIN INFA_REP.OPB_WFLOW_RUN w
ON w.WORKFLOW_RUN_ID=t.WORKFLOW_RUN_ID
WHERE w.WORKFLOW_RUN_ID =
(SELECT MAX(WORKFLOW_RUN_ID)
FROM INFA_REP.OPB_WFLOW_RUN
WHERE WORKFLOW_NAME='wf_emp_load')
ORDER BY t.START_TIME
Output
+---------------+---------------+------------------+-------------------+--------------------+--------------------+
| WORKFLOW_NAME | INSTANCE_NAME | SRC_SUCCESS_ROWS | TARG_SUCCESS_ROWS | START_TIME | END_TIME |
+---------------+---------------+------------------+-------------------+--------------------+--------------------+
| wf_emp_load | s_emp_load | 14 | 14 | 10-JUN-18 18:31:24 | 10-JUN-18 18:31:26 |
| wf_emp_load | s_emp_revert | 14 | 14 | 10-JUN-18 18:31:27 | 10-JUN-18 18:31:28 |
+---------------+---------------+------------------+-------------------+--------------------+--------------------+
|
{
"pile_set_name": "StackExchange"
}
|
// This file is mostly generated by Phil Karn's gensp.c
#include "pch.h"
#ifndef CRYPTOPP_IMPORTS
#include "des.h"
NAMESPACE_BEGIN(CryptoPP)
const word32 RawDES::Spbox[8][64] = {
{
0x01010400,0x00000000,0x00010000,0x01010404, 0x01010004,0x00010404,0x00000004,0x00010000,
0x00000400,0x01010400,0x01010404,0x00000400, 0x01000404,0x01010004,0x01000000,0x00000004,
0x00000404,0x01000400,0x01000400,0x00010400, 0x00010400,0x01010000,0x01010000,0x01000404,
0x00010004,0x01000004,0x01000004,0x00010004, 0x00000000,0x00000404,0x00010404,0x01000000,
0x00010000,0x01010404,0x00000004,0x01010000, 0x01010400,0x01000000,0x01000000,0x00000400,
0x01010004,0x00010000,0x00010400,0x01000004, 0x00000400,0x00000004,0x01000404,0x00010404,
0x01010404,0x00010004,0x01010000,0x01000404, 0x01000004,0x00000404,0x00010404,0x01010400,
0x00000404,0x01000400,0x01000400,0x00000000, 0x00010004,0x00010400,0x00000000,0x01010004},
{
0x80108020,0x80008000,0x00008000,0x00108020, 0x00100000,0x00000020,0x80100020,0x80008020,
0x80000020,0x80108020,0x80108000,0x80000000, 0x80008000,0x00100000,0x00000020,0x80100020,
0x00108000,0x00100020,0x80008020,0x00000000, 0x80000000,0x00008000,0x00108020,0x80100000,
0x00100020,0x80000020,0x00000000,0x00108000, 0x00008020,0x80108000,0x80100000,0x00008020,
0x00000000,0x00108020,0x80100020,0x00100000, 0x80008020,0x80100000,0x80108000,0x00008000,
0x80100000,0x80008000,0x00000020,0x80108020, 0x00108020,0x00000020,0x00008000,0x80000000,
0x00008020,0x80108000,0x00100000,0x80000020, 0x00100020,0x80008020,0x80000020,0x00100020,
0x00108000,0x00000000,0x80008000,0x00008020, 0x80000000,0x80100020,0x80108020,0x00108000},
{
0x00000208,0x08020200,0x00000000,0x08020008, 0x08000200,0x00000000,0x00020208,0x08000200,
0x00020008,0x08000008,0x08000008,0x00020000, 0x08020208,0x00020008,0x08020000,0x00000208,
0x08000000,0x00000008,0x08020200,0x00000200, 0x00020200,0x08020000,0x08020008,0x00020208,
0x08000208,0x00020200,0x00020000,0x08000208, 0x00000008,0x08020208,0x00000200,0x08000000,
0x08020200,0x08000000,0x00020008,0x00000208, 0x00020000,0x08020200,0x08000200,0x00000000,
0x00000200,0x00020008,0x08020208,0x08000200, 0x08000008,0x00000200,0x00000000,0x08020008,
0x08000208,0x00020000,0x08000000,0x08020208, 0x00000008,0x00020208,0x00020200,0x08000008,
0x08020000,0x08000208,0x00000208,0x08020000, 0x00020208,0x00000008,0x08020008,0x00020200},
{
0x00802001,0x00002081,0x00002081,0x00000080, 0x00802080,0x00800081,0x00800001,0x00002001,
0x00000000,0x00802000,0x00802000,0x00802081, 0x00000081,0x00000000,0x00800080,0x00800001,
0x00000001,0x00002000,0x00800000,0x00802001, 0x00000080,0x00800000,0x00002001,0x00002080,
0x00800081,0x00000001,0x00002080,0x00800080, 0x00002000,0x00802080,0x00802081,0x00000081,
0x00800080,0x00800001,0x00802000,0x00802081, 0x00000081,0x00000000,0x00000000,0x00802000,
0x00002080,0x00800080,0x00800081,0x00000001, 0x00802001,0x00002081,0x00002081,0x00000080,
0x00802081,0x00000081,0x00000001,0x00002000, 0x00800001,0x00002001,0x00802080,0x00800081,
0x00002001,0x00002080,0x00800000,0x00802001, 0x00000080,0x00800000,0x00002000,0x00802080},
{
0x00000100,0x02080100,0x02080000,0x42000100, 0x00080000,0x00000100,0x40000000,0x02080000,
0x40080100,0x00080000,0x02000100,0x40080100, 0x42000100,0x42080000,0x00080100,0x40000000,
0x02000000,0x40080000,0x40080000,0x00000000, 0x40000100,0x42080100,0x42080100,0x02000100,
0x42080000,0x40000100,0x00000000,0x42000000, 0x02080100,0x02000000,0x42000000,0x00080100,
0x00080000,0x42000100,0x00000100,0x02000000, 0x40000000,0x02080000,0x42000100,0x40080100,
0x02000100,0x40000000,0x42080000,0x02080100, 0x40080100,0x00000100,0x02000000,0x42080000,
0x42080100,0x00080100,0x42000000,0x42080100, 0x02080000,0x00000000,0x40080000,0x42000000,
0x00080100,0x02000100,0x40000100,0x00080000, 0x00000000,0x40080000,0x02080100,0x40000100},
{
0x20000010,0x20400000,0x00004000,0x20404010, 0x20400000,0x00000010,0x20404010,0x00400000,
0x20004000,0x00404010,0x00400000,0x20000010, 0x00400010,0x20004000,0x20000000,0x00004010,
0x00000000,0x00400010,0x20004010,0x00004000, 0x00404000,0x20004010,0x00000010,0x20400010,
0x20400010,0x00000000,0x00404010,0x20404000, 0x00004010,0x00404000,0x20404000,0x20000000,
0x20004000,0x00000010,0x20400010,0x00404000, 0x20404010,0x00400000,0x00004010,0x20000010,
0x00400000,0x20004000,0x20000000,0x00004010, 0x20000010,0x20404010,0x00404000,0x20400000,
0x00404010,0x20404000,0x00000000,0x20400010, 0x00000010,0x00004000,0x20400000,0x00404010,
0x00004000,0x00400010,0x20004010,0x00000000, 0x20404000,0x20000000,0x00400010,0x20004010},
{
0x00200000,0x04200002,0x04000802,0x00000000, 0x00000800,0x04000802,0x00200802,0x04200800,
0x04200802,0x00200000,0x00000000,0x04000002, 0x00000002,0x04000000,0x04200002,0x00000802,
0x04000800,0x00200802,0x00200002,0x04000800, 0x04000002,0x04200000,0x04200800,0x00200002,
0x04200000,0x00000800,0x00000802,0x04200802, 0x00200800,0x00000002,0x04000000,0x00200800,
0x04000000,0x00200800,0x00200000,0x04000802, 0x04000802,0x04200002,0x04200002,0x00000002,
0x00200002,0x04000000,0x04000800,0x00200000, 0x04200800,0x00000802,0x00200802,0x04200800,
0x00000802,0x04000002,0x04200802,0x04200000, 0x00200800,0x00000000,0x00000002,0x04200802,
0x00000000,0x00200802,0x04200000,0x00000800, 0x04000002,0x04000800,0x00000800,0x00200002},
{
0x10001040,0x00001000,0x00040000,0x10041040, 0x10000000,0x10001040,0x00000040,0x10000000,
0x00040040,0x10040000,0x10041040,0x00041000, 0x10041000,0x00041040,0x00001000,0x00000040,
0x10040000,0x10000040,0x10001000,0x00001040, 0x00041000,0x00040040,0x10040040,0x10041000,
0x00001040,0x00000000,0x00000000,0x10040040, 0x10000040,0x10001000,0x00041040,0x00040000,
0x00041040,0x00040000,0x10041000,0x00001000, 0x00000040,0x10040040,0x00001000,0x00041040,
0x10001000,0x00000040,0x10000040,0x10040000, 0x10040040,0x10000000,0x00040000,0x10001040,
0x00000000,0x10041040,0x00040040,0x10000040, 0x10040000,0x10001000,0x10001040,0x00000000,
0x10041040,0x00041000,0x00041000,0x00001040, 0x00001040,0x00040040,0x10000000,0x10041000}
};
NAMESPACE_END
#endif
|
{
"pile_set_name": "Github"
}
|
Rails.application.configure do
# Settings specified here will take precedence over those in config/application.rb.
# In the development environment your application's code is reloaded on
# every request. This slows down response time but is perfect for development
# since you don't have to restart the web server when you make code changes.
config.cache_classes = false
# Do not eager load code on boot.
config.eager_load = false
# Show full error reports.
config.consider_all_requests_local = true
# Enable/disable caching. By default caching is disabled.
# Run rails dev:cache to toggle caching.
if Rails.root.join('tmp', 'caching-dev.txt').exist?
config.action_controller.perform_caching = true
config.cache_store = :memory_store
config.public_file_server.headers = {
'Cache-Control' => "public, max-age=#{2.days.to_i}"
}
else
config.action_controller.perform_caching = false
config.cache_store = :null_store
end
# Store uploaded files on the local file system (see config/storage.yml for options)
config.active_storage.service = :local
# Don't care if the mailer can't send.
config.action_mailer.raise_delivery_errors = false
config.action_mailer.perform_caching = false
# Print deprecation notices to the Rails logger.
config.active_support.deprecation = :log
# Raise an error on page load if there are pending migrations.
config.active_record.migration_error = :page_load
# Highlight code that triggered database queries in logs.
config.active_record.verbose_query_logs = true
# Debug mode disables concatenation and preprocessing of assets.
# This option may cause significant delays in view rendering with a large
# number of complex assets.
config.assets.debug = true
# Suppress logger output for asset requests.
config.assets.quiet = true
# Raises error for missing translations
# config.action_view.raise_on_missing_translations = true
# Use an evented file watcher to asynchronously detect changes in source code,
# routes, locales, etc. This feature depends on the listen gem.
# config.file_watcher = ActiveSupport::EventedFileUpdateChecker
end
|
{
"pile_set_name": "Github"
}
|
Federal investigators on Thursday recovered a loaded gun that had been smuggled into the notorious Manhattan jail where Jeffrey Epstein killed himself last summer.
The weapon was found at the Metropolitan Correctional Center after a weeklong lockdown that turned up other contraband, including cellphones and drugs, and led to a criminal investigation into guard misconduct, according to the federal Bureau of Prisons.
The handgun was located by Bureau of Prisons officers inside a housing unit at the jail, prison officials said in a statement to the AP.
It marked a massive breach of protocol and raised serious questions about the security practices in place at the Bureau of Prisons, which is responsible for more than 175,000 federal inmates, and specifically at the jail, which had been billed as one of the most secure in America.
Officials have not said where specifically the gun had been found, or how it had been smuggled inside the jail.
Bureau of Prisons officers found a smuggled gun inside a housing unit at the Metropolitan Correctional Center on Thursday, after a six-day lockdown
Meanwhile, federal prosecutors opened a criminal investigation into potential misconduct by jailers, focusing on the flow of contraband into the lockup uncovered during the search for the gun, three people familiar with the matter told the AP.
Attorney General William Barr named a new director last week to take charge of the agency, which has been the subject of intense scrutiny since Epstein, 66, committed suicide while in custody in August. But the agency has been plagued for years by serious misconduct, violence and a chronic staffing shortage
The investigation and search at the jail began last week after officials got a tip that a gun may have been smuggled into the lockup and placed the jail on lockdown 'in order to protect the public, staff and inmates until a comprehensive search could be completed,' the agency said in a statement.
Since then, officials have been keeping inmates locked down in their cells without access to their lawyers and canceled all visitation at the jail, which houses about 700 inmates.
In the past few days, officers have searched the facility and uncovered a sizable amount of contraband, including cellphones, narcotics and homemade weapons, the Bureau of Prisons said.
'All of these items pose a significant threat to the safety and security of the facility as well as the public,' the agency said in a statement.
Investigators were planning to continue searching the jail throughout the night Thursday, looking for any additional contraband, and the lockdown was expected to continue.
Two correctional officers who were responsible for guarding Jeffrey Epstein on the night he hanged himself were allegedly shopping online for furniture and napping
Federal prosecutors are trying to determine how the it has all been entering the facility and the Bureau of Prisons said it notified the Justice Department's inspector general and the FBI.
'The BOP is committed to the safety of staff, inmates and the public while continuing to ensure that those responsible for misconduct and criminal activity are held accountable,' the statement said.
All visitors and inmates are searched before entering the facility and go through metal detectors. They are supposed to leave personal belongings outside the jail. All mail is also screened by correctional staff.
The Bureau of Prisons said the jail has been on 'modified operations' because of the investigation and didn't provide an estimate for when normal operations could resume. Defense attorneys have raised concerns because the jail houses pretrial inmates while their cases are ongoing.
David Patton, executive director and chief attorney of the Federal Defenders of New York, said it's a violation of inmates' constitutional rights to deny them visits with their lawyers. He also said it has affected legal proceedings.
'Sentencings are being delayed. Hearings are being delayed,' Patton said. 'But the MCC acts as though it's perfectly fine for them to just shut down the entire institution to look for contraband. It's just not acceptable. They've got to be able to walk and chew gum at the same time.'
The agency said Thursday that it arranged for additional staff from other parts of the US to assist in the investigation and ensure there is appropriate staffing at the jail.
The agency said it 'has maintained communication with stakeholders as needed' and held a meeting with the chief federal judge and public defender in Manhattan, along with prosecutors, the Marshals Service and probation officials.
The agency also said it has discussed a plan to resume legal visits at the jail on Friday and expects to allow full legal visitation by next week. Visits with family members could resume as soon as this week, officials added.
The Manhattan jail currently houses disgraced attorney Michael Avenatti, who was convicted last month of trying to extort more than $20million from Nike
'The Bureau has been working closely with the stakeholders throughout this period to ensure those defendants with imminent court deadlines have the legal visits with their legal counsel as needed,' the statement said.
Inmates are being locked down for 24 hours a day, and lawyers have been told that on some units the prisoners are being denied showers and being given cold meals in their cells. One inmate reported receiving peanut butter and jelly sandwiches for every meal since last Thursday.
The bureau said inmates are on a periodic shower rotation, except for those in special housing units, who remain on a regular schedule.
'All inmates have access to medical care and appointments and medical staff continue normal rounds on every floor,' the Bureau of Prisons said.
It is just the latest crisis at the jail, which houses a number of high-profile inmates, including attorney Michael Avenatti, who gained fame by representing porn actress Stormy Daniels in lawsuits involving President Donald Trump.
Federal prosecutors allege that the two correctional officers assigned to watch Epstein's unit were snoozing and shopping for furniture on the internet when he hanged himself in his cell on August 10, and later forged records to make it look like they checked in on him.
After they discovered the high-profile inmate dead, the officers allegedly told a supervisor they had 'messed up' and 'didn't do any checks' in the hours before he killed himself.
Jeffrey Epstein hanged himself on August 10 in his cell at the Metropolitan Correctional Center while awaiting trial on federal sex trafficking charges
The two guards are accused of repeatedly signing false certifications saying that they had conducted multiple counts of inmates during their shift.
The prisoners were not checked on for eight hours, according to the indictment. The guards discovered Epstein's body at 6.30am.
Epstein hanged himself on August 10 in his cell while awaiting trial on federal sex trafficking charges.
His death has triggered investigations by the FBI, the US Department of Justice's Office of Inspector General and the U.S. Bureau of Prisons, which runs the detention facility.
It also prompted outrage and disbelief over how such a high-profile prisoner, known for socializing with powerful people including presidents Donald Trump and Bill Clinton and Prince Andrew, could have been left without surveillance at the federal facility.
Epstein's suicide occurred despite a previous attempt on July 23 when he was found on the floor of his cell with bruises on his neck.
|
{
"pile_set_name": "OpenWebText2"
}
|
Background {#Sec1}
==========
Mycobacterium species represent both environmental and pathogenic organisms that fall into two major groups: tuberculosis complex such as *M. tuberculosis* and *M. bovis* (MTBC), and Non-tuberculous mycobacteria (NTM) such as *M. avium* complex, *M. abscessus* and *M. smegmatis*. It is estimated that across the world 9.6 million people are infected with tuberculosis every year, 3.6 million of these people are not given proper treatment, and 1.5 million people die from infection \[[@CR1]\]. NTM infections have become a growing concern as more people with lung infections have positive cultures for NTM species \[[@CR2]\], with cystic fibrosis patients representing a disproportionate amount of detected infections \[[@CR3]\]. The prevalence of NTM disease, while relatively rare at 86,244 cases in 2010 in the United States \[[@CR4]\], is increasing throughout the world \[[@CR5], [@CR6]\], with incidence of NTM exceeding that of tuberculosis in the United States \[[@CR6]\]. Treatment of NTM disease also presents a problem due to the chronic nature of the disease, antibiotic treatments lasting up to 18 months, and the cost of treatment being higher than that of multi-drug resistant tuberculosis \[[@CR4]\]. Better understanding of gene function for these species allows for better interpretation of clinical experiments, leading to an increased understanding of gene roles and potential drug targets.
Predictive functional annotation methods are a standard practice in analyzing genome sequencing data \[[@CR7]\]. Current gene annotation and protein functional annotations are the result of both manual curation and prediction based upon machine-learning tools such as GenemarkS \[[@CR8]\], RAST \[[@CR9]\] and various homology-based methods such as FASTA \[[@CR10]\]. Over the past few years, there has been a development of methods that take into account orthology, protein-protein interactions, and text mining, such as eggNOG \[[@CR11], [@CR12]\], a tool used to better annotate the M. *tuberculosis* genome. There have also been improvements to homology-based methods, allowing for both improved accuracy and the assigning of GO terms to genes \[[@CR13]\]. Improvements in common methodology for annotation prediction has allowed for both better understanding of genomic content and improved analyses performed on genomic and transcriptomic data.
While there are a couple of well curated databases for *M. tuberculosis* data through TBDB \[[@CR14], [@CR15]\] and TubercuList \[[@CR16]\] and a database devoted to *M. abscessus* in MabsBASE \[[@CR17]\], there remains a lack of well-curated databases for mycobacterium genomes as a whole. One early attempt to fill this gap was made by GenoMycDB \[[@CR18]\], a collection of six mycobacterial genomes; however, this database has not been updated to include more genomes. TubercuList was later extended into MycoBrowser \[[@CR19]\]. This website contains a comprehensive genomic and proteomic database for three additional mycobacterial species; although, it still lacks commonly studied NTM such as *M. avium* complex and *M. abscessus*. While TBDB \[[@CR14], [@CR15]\] has grown to include other NTM species, annotations for these species remain limited. PATRIC \[[@CR20]\] contains a wide array of annotated genomes, including mycobacteria, however their functional annotations do not perform well for genomes with large amounts of pseudo genes such as *M. leprae*, leading to 3607 extra genes being annotated despite validation of these as pseudogenes \[[@CR21]\]. The MycoBASE database was created to extend the functional annotation knowledge of mycobacteria in general, allowing for a better genomic understanding of both a highly prevalent group of infectious agents, tuberculosis complex, and a group of emerging pathogens, NTM.
Construction and content {#Sec2}
========================
Mycobacteria gene data {#Sec3}
----------------------
Functional reannotation, gene ontology (GO), phage, and transposon annotation was performed on genes of 11 tuberculosis complex and NTM genomes, as shown in Table [1](#Tab1){ref-type="table"}. Predicted open reading frames (ORFs) from these genomes and their functional annotations were downloaded from two different sources: NCBI and TBDB \[[@CR14], [@CR15]\] and form the standard for our reannotation efforts.Table 1Mycobacterial database speciesSpeciesStrainAbbreviationSourceGenesReference*M. abscessus sub. abscessus*ATCC 19977MABTBDB4942\[[@CR40]\]*M. tuberculosis*H37RvMTBNCBI4018\[[@CR16]\]*M. bovis*AF2122/97MBOVISTBDB3920\[[@CR41]\]*M. avium*104MAVTBDB5120\[[@CR42]\]*M. abscessus sub. massiliense* ^a^CRM0020CRMNCBI4750\[[@CR43]\]*M. abscessus sub. massiliense* ^a^CCUG48898MMASNCBI5193\[[@CR44]\]*M. abscessus sub. bolletii*CIP108541MBOLNCBI4923\[[@CR45]\]*M. leprae*TN1MLEPRAETBDB1605\[[@CR21]\]*M. intracellulare*ATCC 13950MINTNCBI5144\[[@CR46]\]*M. kansasii*ATCC 12478MKANNCBI5449unpublished*M. smegmatis*MC2 155MSMEGTBTB6716unpublishedSpecies in the initial release of the database that contain both functional and GO term annotation data. ^a^ Subspecies *massiliense* is currently listed as subspecies *bolletii* in NCBI; however, the classification is still being debated and the distinction between *massiliense* and *bolletii* has clinical importance \[[@CR47]\]
Annotation prediction {#Sec4}
---------------------
Blannotator \[[@CR13]\] was chosen for functional reannotation and GO term annotation due to its high accuracy for bacterial genomes relative to other homology-based methods. Blannotator source code \[[@CR13]\], UniProt databases \[[@CR22]\], GO databases \[[@CR23]\], and NCBI-BLAST+ libraries \[[@CR24]\] were downloaded onto a linux server. Blannotator source code was then modified to utilize NCBI-BLAST+ libraries and to utilize more threads when running BLAST+. Predicted ORFs from each of the 11 genomes were then translated and annotated with Blannotator (improved annotations). Only the highest scoring functional annotations were used for all downstream assessments and database generation. All GO terms associated with a protein were used due to both the multiple associations between GO terms and a given function, and the hierarchical structure of GO terms. To annotate genes associated with DNA transposons, we used the predicted GO term "transposase activity". Genes associated with phage regions were annotated with PHAST \[[@CR25]\] due to its high accuracy compared to other phage tools. Both phage and transposon are annotated with "YES" or "NA" to denote their likelihood of being part of a phage or transposon region. The output of all of these annotations were then stored in databases described in the Database Structure section with the pipeline for creating these annotations seen in Fig. [1](#Fig1){ref-type="fig"}.Fig. 1Pipeline for creating improved annotations. Input files were taken from NCBI and TBDB and then annotated using Blannotator and PHAST. The resulting annotations were then used to create databases containing GO terms and a collection of functional, phage, and transposon annotations
Evaluation of improved functional annotations {#Sec5}
---------------------------------------------
For a protein to be considered annotated, the protein must meet at least one of the three characteristics: function, localization, and/or name. Examples of each of the following are as such: "methyltransferase", "membrane protein", and "fadE6". Terms such as "precursor" and "10.1 kDa protein" are excluded, as they do not represent any of these characteristics. Uncharacterized proteins that have protein names are considered annotated as they often make up large families of homologous proteins that have uncertain functions, such as PPE family proteins and UPF/DUF proteins. Uncharacterized proteins where that annotation matches the protein ID are not considered annotations. This annotation guideline conforms to other reannotation ventures such as EggNOG \[[@CR11]\], as well as giving consistent coverage of terms across the original and improved annotations.
Preprocessing was performed on annotations to allow for accurate comparison of function, localization, and name. Non-alphanumeric characters were replaced by blanks due to their inconsistent use in separating words, compounding words, and naming of chemical entities. Words that represent homology scoring or redundancy in naming, such as "putative", "family", "protein", etc. are also removed for this evaluation; however, these annotations are maintained in the database.
Overlapping annotations from the original and improved annotations are then compared using bigram Dice's coefficient, a coefficient used in natural language processing to compare word sets \[[@CR26]\]. Bigrams, sets involving two successive letters, are used to preserve some of the lexicon that was eliminated by removing non-alphanumeric characters. Given the structured vocabulary present within both original and improved annotations, Dice's coefficients offer a precise method for automated comparison of annotations. To identify if two annotations are significantly similar, a Wilcoxin signed rank test with a confidence interval of 95 % was used. The background for the test was the set of Dice's coefficients generated by comparing original annotations against all other original annotations. An annotation pair between the original and improved was considered significantly similar if their Dice's coefficient was above the 95 % confidence interval. Reasons for insignificance include: generic annotations, same function but different annotation, similar but different functions, and completely different function.
Each of the 11 Mycobacterial genomes was evaluated for functional annotation and GO term annotation coverage. While none of the 11 genomes had GO terms to evaluate against, all the genomes had functional annotations. Blannotator produced an average increase in functional annotations for each genome of 20 %, shown in Fig. [2a](#Fig2){ref-type="fig"}, ranging from 11 % in CRM to 31 % in MINT. In addition to this increase in functional annotations, the average coverage of GO terms was 9 % higher than the original functional annotations. The average coverage for GO terms is 75 % of genes, ranging from 71 % in MMAS to 82 % in MLEPRAE. This results in a significant increase in functional annotations, in addition to having GO terms for functional enrichment testing. Figure [2b](#Fig2){ref-type="fig"} shows the percent of annotations that overlapped between the original and improved annotations. This figure also shows the percentage of overlapping annotations that were significantly similar relative to the background. The result of this evaluation showed that an average of 99.6 % of annotations overlapped (range: 97.6 % in MKAN to 100 % in MBOL) and that 93.1 % of these overlapping annotations were significantly similar (range: 89.4 % in CRM to 97 % in MMAS).Fig. 2Annotation improvement in the 11 mycobacterial genomes. **a** Proportion of genes with original functional annotations (*black*), GO term annotations (*green*), and improved functional annotations (*blue*). **b** Proportion of original functional annotations with a corresponding improved functional annotation (*black*), and the proportion of these annotations that are more significantly similar than background (*red*)
Gene ontology enrichment {#Sec6}
------------------------
Modified one-sided Fisher's exact tests, similar to those created for EASE scores \[[@CR27]\], are used to evaluate enrichment of GO terms in a gene set against a background set. A hypergeometric probability for contingency tables is calculated using an estimation \[[@CR28]\], allowing for a more efficient calculator than the direct representation of the Fisher's exact test. For calculating *p*-values, the GO terms had to meet two criteria: the number of genes associated with the GO term in the gene set is greater than one, and the proportion of genes with the GO term is greater in the gene set than in the background set. All genes in the gene set and the background set, irrespective of whether or not they have GO terms associated with them, count towards the values in the contingency tables. Both the non-multiple testing corrected *p*-values and Bonferroni adjusted *p*-values are ranked and displayed. This Java-based program is available for download and use on the Website.
Database structure {#Sec7}
------------------
The database is made up of two tables, as shown in Fig. [3](#Fig3){ref-type="fig"}. The feature table contains all known gene information for a given genome. This contains the strain identifier, the gene ID, the common name from NCBI/TBDB, the location of the gene, whether or not the gene is related to a transposon or phage, the original functional annotation, and the improved functional annotation. This table can be queried by selecting a genome of interest and by either selecting all genes or a supplying a subset of genes. The GO table contains GO information for all genes that contain GO Terms. This GO information contains the ID, the term, and the namespace of a GO term for a given gene. Each gene can contain multiple GO terms. This database is queried through the first table, as these tables are linked together by gene IDs.Fig. 3Database structure. Database containing two tables. The first table contains gene features for a given genome. The second table contains GO terms for all genes. These two tables are linked together by gene IDs
Website {#Sec8}
-------
The database can be accessed from the website: strong.ucdenver.edu/mycobase. From the homepage, users can access pages to search annotations, search GO terms, view a list of currently annotated genomes, and access quick help about using the webpage. On the annotation page, the user first selects their genome of interest from the drop down menu. After selecting the genome of interest, the user selects an option button corresponding to 1) Downloading the whole genome, 2) Searching by gene names, or 3) Searching by annotation. If the user selects search by gene name or annotations, they enter either a single gene/annotation or a list of genes/annotations (separated by comma or newline) into the text box. An example of searching by gene name in the *Mycobacterium tuberculosis* H37Rv genome is "Rv0001". An example of searching by annotation is "methyltransferase". Clicking submit downloads a formatted file of genes corresponding to the genome, gene name, or annotation. The header for describing the formatted information is the first line in the file. A simple flow through of downloading annotations corresponding to gene IDs can be seen in Fig. [4](#Fig4){ref-type="fig"}.Fig. 4Extracting annotations by gene name from website. To download annotations by gene name first click on the "Annotation" link on the website. Next select your genome of interest from the species dropdown menu. Select the "Search gene names" option button then insert a list of gene names separated by a comma or newline character. Next hit the "Submit" button and the list of annotations associated with the gene names will be downloaded
Searching for GO terms follows a similar format as searching annotations. A user first selects the genome of interest, and then selects the option box associated with either the genome or search by gene id. If searching by gene id, the user inputs either a single gene id or list of gene ids separated by comma or newline. Clicking "submit" downloads the list of GO terms with the first line being the header describing each field. The enrichment program for the modified fisher's exact test and a use case is included on this page. A description of the required input files and program description are also included in this download. In addition to being able to enrich for GO terms, this program can also enrich for any categorical terms that can meet the input file guidelines, such as other available *M. tuberculosis* categorical terms \[[@CR29]\]. Lastly, the help page briefly describes how to download GO Terms and annotations.
Utility and discussion {#Sec9}
======================
Exploration in genome variability {#Sec10}
---------------------------------
Predicted genes from MAB and MBOL were compared for sequence homology to differentiate between shared and unique genes between two *Mycobacterium abscessus* genomes. From this analysis we have discovered a 37KB insertion sequence in *Mycobacterium abscessus* ATCC19977, as seen in Fig. [5](#Fig5){ref-type="fig"}. Using predicted GO terms for both of these genomes and the Java-based enrichment program, we have found that this insertion sequence contains a cassette of 8 genes associated with biphenyl and aromatic hydrocarbon degradation enzymes, including a group of ferredoxin reductases that are necessary for iron-catalyzed hydroxylation \[[@CR30], [@CR31]\]. These enzymes allow for degradation of carbon sources such as plant lignin, crude oil, and natural gases, and environmental contaminants such as petroleum products, PCBs, and PAHs. This degradation activity has been observed in a variety of environmental microbes including mycobacteria \[[@CR30]--[@CR34]\]. This style of analysis has been used to analyze content of deletions in *Mycobacterium abscessus* \[[@CR35]\].Fig. 5Insertion of gene cassette in *Mycobacterium abscessus* strain ATCC 19977. Insertion of a 37KB insertion region encoding for a cassette of eight biphenyl and aromatic hydrocarbon degradation enzymes (*red arrows*). Conserved genes within MAB and MBOL are shown in *green*, and 34 inserted genes are shown in *blue* and *red*
Gene ontology term taxonomy {#Sec11}
---------------------------
Creating GO term taxonomy allows for identifying both conserved function across multiple mycobacterium and identifying species-specific functions. Slightly more than half of GO terms are shared by the 11 species, with 73 % of terms being shared by more than half of the species, as shown in Table [2](#Tab2){ref-type="table"}. This shows that the majority of function is conserved across mycobacterium species. Of the GO terms associated with MLEPRAE, 91 % of them occur in the 10 other species, suggesting that MLEPRAE contains a fundamental set of functions that define the mycobacterium species. Only 12 % are unique to a single species, with MSMEG accounting for 62 % of one taxa terms (Fig. [6](#Fig6){ref-type="fig"}). Much of MSMEG's unique terms are carbon-based metabolism and synthesis related, suggesting that its larger genome size allows it to both utilize and create additional carbon sources relative to other mycobacterium \[[@CR36]\]. However, 61 % of GO terms for MSMEG are still shared across all mycobacterium. While MSMEG has more genes and one taxa GO terms, most of the function within MSMEG is conserved across species suggesting that the increased genome size is due to gene duplication \[[@CR37]\]. MTB and MBOVIS share the most 2 taxa terms with 25 % of the total, owing to their similar genome size and their pathogenesis.Table 2Gene ontology term taxonomyTaxaTermsPercent154511.8 %21713.7 %31232.7 %41753.8 %52104.6 %61463.2 %7972.1 %8891.9 %9671.5 %1055912.1 %11242852.7 %Unique GO terms for the intersection of the 11 Mycobacterium species. More than half (52.7 %) of GO terms are shared by all speciesFig. 6Gene ontology 1 taxa terms. Proportion of the 545 GO terms that are unique to one genome. MSMEG contains the majority of 1 taxa terms, owing to its larger genome and diversity of chemicals that it can metabolize and synthesize
Gene ontology term enrichment between genomes {#Sec12}
---------------------------------------------
To evaluate characteristics of mycobacteria, GO term enrichments were performed on select mycobacteria, as shown in Tables [3](#Tab3){ref-type="table"} and [4](#Tab4){ref-type="table"}. Backgrounds for these comparisons were the combination of both of the genomes being compared. The *M. abscessus* and *M. tuberculosis* clades acted as controls for the enrichments due to their similar phenotypes and pathogenicity. Upon analysis, there were no enriched GO terms within these sets, affirming the similarity between the genomes and the validity of the method. The other genome pairs represent differences in growth, pathogenicity and clade. Figure [7](#Fig7){ref-type="fig"} shows the ratio of genes associated with enriched GO terms between the genome pairs. While enriched GO terms in MLEPRAE had lower gene ratios than other genomes, there were a higher proportion of these genes in the genome. This suggests that not all GO functions scale with genome size and that genome size differences are an important consideration when performing enrichments.Table 3GO term enrichment genomesGenomeGenesGrowthTypeCladeMBOV3920SlowObligate PathogenMTBCMLEP1605SlowObligate PathogenUngroupedMTB4018SlowObligate PathogenMTBCMKAN5449SlowEnvironmental-Opportunistic PathogenKansasiiCRM4750FastEnvironmental-Opportunistic Pathogen*M. abscessus* groupMAB4942FastEnvironmental-Opportunistic Pathogen*M. abscessus* groupMBOL4923FastEnvironmental-Opportunistic Pathogen*M. abscessus* groupMMAS5193FastEnvironmental-Opportunistic Pathogen*M. abscessus* groupMSMEG6716FastEnvironmentalUngroupedThe number of genes and general phenotypic traits of the genomes used for the comparisonTable 4Gene enrichment comparisonsGenome1Genome2Gene differenceEnriched GO termsTypeMABMSMEG17741-35DifferentMABMKAN50752-1DifferentMSMEGMLEP511115-270DifferentMTBMAB92448-15DifferentMTBMSMEG269874-37DifferentMTBMLEP24130-26DifferentMTBMKAN143159-0DifferentMABMMAS2510-0SameMABMBOL190-0SameMABCRM1920-0SameMMASMBOL2700-0SameMTBMBOV980-0SameBoth genomes were compared against each other. The gene difference is the difference in the total number of genes between genomes. Enriched GO terms are the number of enriched terms in Genome1 over Genome2 "-" Genome2 over Genome1. If the type is different, then there was either a difference in pathogenicity, slow or fast grower, or the genomes come from different cladesFig. 7Ratio of genes in enriched GO terms. Plot of the 10 genome pairs with more than one enriched gene. The ratio is the number of genes with a given GO term in one genome over the number of genes with that same GO term in the other genome
Enrichment of host-pathogen GO terms occurred in all of the pathogen-environment comparisons except MTB-MKAN; however, in this pair, these terms barely failed to meet significance (*p*-value \~0.07), suggesting these terms are still likely an important distinction between the pair. MSMEG had enrichments in carbohydrate transporters over both MAB and MTB. This is related to the fact that MSMEG can metabolize a broader range of carbohydrates relative to other mycobacterium \[[@CR36]\] and is supported by the number of one taxa GO terms related to carbon metabolism.
Conclusion {#Sec13}
==========
MycoBASE currently contains 11 mycobacterial genomes with functional and GO term annotations. Our genomes are based off of NCBI gene annotations, allowing for a well-accepted genome leading to consistency across studies. These annotations allow for improved understanding of the genetic content of these genomes, leading to more coverage in analyses involving differential gene content (insertion/deletion of genes, differences in genes across species), genes that are understudied but have homology to genes of known function, and functional analyses of transcriptomics and genomics data using GO terms (the modified Fisher's program being available for download on our server). These annotations will be available for download, allowing for a wide variety of analyses. Our future plans include adding a greater diversity of genomes to our database, such as *M. africanum* \[[@CR38]\], *M. chelonae* \[[@CR39]\], and other studied mycobacteria, greatly increasing the number of species in the database.
Availability and requirements {#Sec14}
=============================
This database and GO enrichment is available for academic and other non-commercial uses at the website: strong.ucdenver.edu/mycobase.
GO
: Gene Ontology
MAB
: *M. abscessus sub. abscessus* ATCC19977
MTB
: *M. tuberculosis* H37Rv
MBOVIS
: *M. bovis* AF2122/97
MAV
: *M. avium* 104
CRM
: *M. abscessus sub. massiliense* CRM0020
MMAS
: *M. abscessus sub. massiliense* CCUG48898
MBOL
: *M. abscessus sub. bolletii* CIP108541
MLEPRAE
: *M. leprae* TN1
MINT
: *M. intracellulare* ATCC13950
MKAN
: *M. kansasii* ATCC12478
MSMEG
: *M. smegmatis* MC2 155
**Competing interests**
The authors declare that they have no competing interests.
**Authors' contributions**
BG- Conceived project, methods development, coding, paper writing. GD- Webpage development, methods development, paper editing. RD- Methods development, paper editing. MS- Conceived project, oversaw project, paper editing. All authors read and approved the final manuscript.
BG acknowledges support from a NIH Biomedical Informatics training grant 2T15LM009451-06; RD and MS acknowledge support from the National Jewish Health NTM Center of Excellence funded in part by the Amon G. Carter Foundation; M.S. acknowledges support from the Colorado Bioscience Discovery Program, the Eppley Foundation, and the Boettcher Foundation Webb-Waring Biomedical Research Program. We thank Sonia Leach for assisting with the setup and installation of software and libraries and David Farrell for helping set up the website server.
|
{
"pile_set_name": "PubMed Central"
}
|
Former domestic intelligence chief Hans-Georg Maaßen and Interior Minister Horst Seehofer | Bernd Von Jutrczenka/AFP via Getty Images Seehofer popularity drops amid intelligence chief scandal: poll Angela Merkel’s conservative bloc also took a hit in the wake of Hans-Georg Maaßen’s removal, survey finds.
German Interior Minister Horst Seehofer's approval ratings have taken a hit in the aftermath of a scandal over the country's intelligence chief, according to a new poll published Friday.
About 59 percent of Germans participating in a poll by German public broadcaster ARD said they believed Seehofer was "a bad choice" to lead the interior ministry. Only 28 percent said they supported Seehofer, a drop of 11 percentage points since April.
The low for Seehofer, who is also the leader of Angela Merkel's Bavarian allies, the Christian Social Union (CSU), comes just weeks ahead of a Bavarian election in which his party — which has dominated the wealthy southern state’s politics since World War II — looks likely to lose its absolute majority.
Seehofer came under fire in recent weeks for backing intelligence chief Hans-Georg Maaßen when he came under attack for questioning the authenticity of a video showing a foreigner being chased at a far-right rally in Chemnitz, as well as allegations he provided the anti-immigrant Alternative for Germany (AfD) party with confidential government material.
Merkel's governing coalition removed Maaßen from his job earlier this week, but granted him a better-paid, more senior position as state secretary in Seehofer's ministry.
The heavy blowback over the decision also translated into a drop in support for the chancellor's conservative bloc, according to the new poll, which found that only 28 percent of respondents said they would cast their vote for her alliance if an election were held on Sunday. The result marks the bloc's worst performance in the ARD poll since 1997.
The AfD became the country's second most popular party after Merkel's conservatives, polling at 18 percent, one percentage point above the center-left Social Democrats. The far-right party overtook the SPD for the first time in a national poll in February earlier this year.
|
{
"pile_set_name": "OpenWebText2"
}
|
Help. Recording into Creative Live Drive
I have been practicing my harp techniques for a few months now and would like to record my practice sessions into my computer through my Creative Sound Blaster Live drive. There are two MIDI ports (one in, one out) as well as SPDIF ports in and out.
Microphone aside, do I need to purchase a device, such as a mixer, to feed into the Live Drive or can it be done directly from the mic. I have not purchased the mic yet but I am leaning towards the Green Bullet.
|
{
"pile_set_name": "Pile-CC"
}
|
Q:
How to limit broadcast to its own Android app
I wondering if it is possible to create a private broadcast.
I actually register broadcastreceiver in my activity and use
sendOrderedBroadcast(broadcast);
method to send the broadcast.
But, for now, the intent (broadcast) used for this method is defined like this :
Intent broadcast = new Intent("com.mypackage.broadcast");
So every external application which declare this package name can listen what I am sending, and I don't want to.
So how to make this impossible, that no one can listen my broadcast ?
A:
I think you are looking for LocalBroadcast Manager. The docs say:
It is a helper to register for and send broadcasts of Intents to local objects within your process. This is has a number of advantages over sending global broadcasts with sendBroadcast(Intent). One of them is that the data you are broadcasting won't leave your app, so don't need to worry about leaking private data.`
See how to use LocalBroadcastManager? for more. Hope it helps you.
|
{
"pile_set_name": "StackExchange"
}
|
Reddit Flip Pin 107 Shares
The future of Charlie Cox’s Matt Murdock might be up in the air following Netflix’s cancellation of Daredevil, but we now have confirmation that will be seeing more of Karen Page, with Deborah Ann Woll set to reprise her role in season two of The Punisher.
“We were shooting while Daredevil was shooting [its third season],” showrunner Steve Lightfoot Deborah has a huge part in that season,” Lightfood told Screen Rant when asked if Karen Page will reunite with Jon Bernthal’s Frank Castle in the new season. “Deborah was very busy, and I didn’t know when I could get her into the show this time around. In the world of the show, that relationship [between Frank and Karen] continues to be very important.”
Based on Lightfoot’s comments, it doesn’t sound like Karen will play too significant a role in the new season of The Punisher, which – based on Netflix’s recent Marvel track record – will likely be its last, on the streaming service at least.
SEE ALSO: The Punisher season 2 coming to Netflix in January
The second season of The Punisher will see Jon Bernthal returning as Frank Castle alongside Ben Barnes as Billy Russo, Amber Rose Revah as Dinah Madani, Jason R. Moore as Curtis Hoyle, Josh Stewart as John Pilgrim, Floriana Lima as Krista Dumont, Giogia Whigham as Amy Bendix, Annette O’Toole as Eliza Schultz, and Corbin Bernsen as Anderson Schultz.
|
{
"pile_set_name": "OpenWebText2"
}
|
M & S / Off Road / Aggressive treads
This section is for any vehicles equipped with
large, aggressive tires, similar to those made by Interco, BFGoodrich, Toyo, Mickey Thompson, and PitBull, although the data can be applied to any manufacturer
of similarly sized tires. Make sure you check the "Ply"
column when making your selection.
Note: This chart was developed using a cross-section of data supplied by various manufacturers. Some tires may need different amounts depending on the individual tire.
If your size is not listed here, email us and we'll supply the correct amount.
Tire Width
Aspect Ratio
Rim Dia
Ply Rating
Ounces
LT215
70
15
C
5
LT215
75
15
C
5
LT225
75
16
C
5
LT235
70
16
C
6
LT235
75
15
C
5
LT235
85
16
E
6
LT245
70
16
D
6
LT245
75
16
E
6
LT255
70
16
D
6
LT265
70
16
D
6
LT265
75
16
D
6
LT275
70
16
D
6
LT285
75
16
D
8
LT285
75
16
E
8
LT295
75
16
D
8
LT305
70
16
D
8
LT315
75
16
D
8
LT315
75
16
E
8
LT255
85
16
C
6
LT245
70
17
E
6
LT265
70
17
D
6
LT285
70
17
D
6
LT305
65
17
E
8
LT315
70
17
D
8
290
70
18
D
8
Continental / Size
Rim Dia
Ply Rating
Ounces
325 MPT
70
18
8
12
365 MPT
70
18
8
12
275 MPT
80
20
10
10
335 MPT
80
20
16
14
365 MPT
80
20
14
20
365 MPT
80
20
16
20
405 MPT
70
20
16
20
455 MPT
70
24
18
20
Continental / Size
Rim Dia
Ply Rating
Ounces
10.5 MPT E5
18
10
8
12.5 MPT 30
18
12
12
10.5 MPT MIL
20
10
8
10.5 MPT Titan
20
14
10
10.5 MPT 80
20
10
10
10.5 MPT 80
20
14
10
12.5 MPT 30
20
12
12
12.5 MPT E6
20
12
10
12.5 MPT 80
20
12
12
12.5 MPT 80
20
16
14
12.5 MPT 80
20
22
14
14.5 MPT E6
20
10
14
14.5 MPT 30
20
10
14
14.5 MPT E6
20
18
14
14.5 MPT 30
20
18
14
14.5 MPT 80
20
10
14
14.5 MPT 80
20
18
16
14.5 MPT 80
20
22
16
14.5 MPT 70E
20
16
20
14.5 MPT 81
20
14
20
Michelin / Size
Tread
Load Range
Ounces
9.00 R16
XL
D
8
11.00 R16
XL
E
10
14.75/80 R20
XL
J
22
15.5/80 R20
XL
J
25
11.00 R20
XL
H
14
12.50 R20
XL MPT
F
12
14.00 R20
XL
J
25
14.5 R20
XL
L
16
16.00 R20
XL
J
36
11.00 R20
XZL
H
16
14.00 R20
XZL
J
26
16.00 R20
XZL
M
36
365/85 R20
XZL
J
21
395/85 R20
XZL
J
24
335/80 R20
XZL
F
12
365/80 R20
XZL
L
18
24 R21
XZL
H
45
14.00 R20
XZL
J
40
16.00R20
XZL
J
72
325/85 R16
XML
D
14
12.00 R20
XML
J
20
395/85 R20
XML
G
24
14.00 R20
XML
G
30
Tire Dia
Rim width
Rim Dia
Ply Rating
Ounces
27
9.5
14
6
6
27
9.5
15
6
5
28
8.5
14
4
4
29
8.5
15
4
4
29
10.5
15
4
5
29
10.5
15
6
6
29
11.5
15
6
6
30
10.5
15
6
5
30
11.5
15
4
6
30
11.5
15
6
8
30
11.5
16
6
6
31
9.5
15
6
5
31
10.5
15
6
6
31
11
15
6
8
31
11
16
6
6
31
11.5
15
6
6
31
11.5
16
8
6
31
12.5
15
6
8
31
12.5
16
6
6
31
12.5
16
10
8
31
13.5
15
6
8
32
9
16
6
6
32
9
16
10
6
32
9.5
15
6
6
32
10
16
6
6
32
10.5
15
6
6
32
11.5
15
6
8
32
11.5
16
6
8
32
11.5
16
6
8
32
11.5
16
10
8
33
10.5
15
6
6
33
10.5
15
6
8
33
10.5
16
8
6
33
11.5
15
6
8
33
11.5
16
8
8
33
11.5
16.5
8
8
33
12.5
15
6
8
33
12.5
16
6
6
33
12.5
16
10
8
33
12.5
16.5
6
8
33
12.5
16.5
10
10
33
12.5
17
10
8
33
12.5
18
10
8
33
12.5
20
10
8
33
13
16
8
8
33
13.5
15
6
8
33
13.5
16
8
8
33
13.5
17
8
8
33
13.5
18
8
8
33
13.5
20
8
8
33
14
15
6
8
33
14
16
6
8
33
14
16.5
6
8
33
14.5
15
6
8
33
14.5
16
10
10
33
14.5
16.5
10
10
33
15.5
15
6
10
33
15.5
16.5
6
8
33
15.5
16.5
6
8
34
9
16
6
6
34
9.5
16.5
6
6
34
10.5
16
8
8
34
10.5
17
8
8
34
12.5
15
6
8
34
12.5
16
10
8
35
10.5
15
6
8
35
10.5
16
10
8
35
10.5
16
8
8
35
12.5
15
6
8
35
12.5
16
8
8
35
12.5
16
10
8
35
12.5
16.5
6
8
35
12.5
17
10
8
35
12.5
18
10
8
35
12.5
20
10
8
35
13.5
15
6
8
35
14.5
15
6
8
35
14.5
16
6
8
35
14.5
16
10
8
35
14.5
16.5
10
8
35
14.5
16.5
6
8
35
14.5
17
8
8
35
14.5
18
8
8
35
14.5
20
8
8
35
15
15
6
8
35
15
16.5
10
8
35
15
16.5
6
8
35
15.5
15
6
8
35
15.5
15
6
10
35
15.5
16.5
6
10
35
16
15
6
8
35
16
16
6
8
35
16
16.5
6
10
35
16
16.5
10
10
36
12.5
15
8
8
36
12.5
15
6
10
36
12.5
15
6
10
36
12.5
15
6
8
36
12.5
16
10
8
36
12.5
16.5
6
8
36
12.5
16.5
8
8
36
13
16
6
8
36
13.5
15
6
10
36
13.5
15
6
10
36
13.5
16
10
10
36
13.5
16
8
8
36
13.5
16.5
10
10
36
13.5
16.5
8
8
36
13.5
17
10
10
36
13.5
17
8
8
36
13.5
20
10
8
36
13.5
20
8
8
36
14.5
15
6
10
36
14.5
16
10
10
36
14.5
16.5
10
10
37
12.5
15
6
8
37
12.5
16
10
10
37
12.5
16.5
10
10
37
12.5
17
10
10
37
12.5
20
10
10
37
13
15
6
8
37
13
16
6
8
37
13
16.5
6
8
37
13
16.5
8
8
37
13
17
6
8
37
13.5
17
10
10
37
13.5
18
10
10
37
13.5
20
10
10
37
13.5
22
10
10
37
13.5
24
10
10
37
14
15
6
8
37
14
16
8
8
37
14
16.5
8
8
37
14.5
15
6
8
38
12.5
15
6
10
38
12.5
16.5
6
10
38
12.5
16.5
10
10
38
13
16
6
10
38
14.5
16
8
12
38
15.5
15
6
10
38
15.5
16
10
12
38
15.5
16.5
10
12
38
15.5
17
10
10
38
15.5
18
10
10
38
15.5
20
10
10
38.5
11
15
6
10
38.5
11
16
6
10
38.5
11
16.5
6
10
38.5
14.5
15
6
12
38.5
14.5
16
6
12
38.5
14.5
16.5
6
12
38.5
15
15
6
10
38.5
15
16
6
10
38.5
15
16.5
6
10
38.5
16
15
6
10
38.5
16
16.5
6
10
39.5
13.5
15
6
12
39.5
13.5
16
10
12
39.5
13.5
16
8
10
39.5
13.5
16.5
10
12
39.5
13.5
16.5
8
10
39.5
13.5
17
10
12
39.5
13.5
17
8
10
39.5
13.5
20
10
10
39.5
13.5
20
8
8
39.5
15
15
6
12
39.5
15
16
6
12
39.5
15
16.5
10
12
39.5
15
16.5
6
10
39.5
15
16.5
8
12
39.5
15
17
6
12
39.5
15
20
8
10
39.5
18
15
6
12
39.5
18
16.5
6
12
40
16
15
6
12
40
15.5
22
8
10
41
14.5
16
8
12
41
14.5
17
8
12
41
14.5
18
8
12
41
14.5
20
8
10
41
14.5
22
8
10
42
14
15
6
12
42
14
16
6
10
42
14
16.5
6
12
42
14
17
6
12
42
15
15
6
12
42
15
16.5
10
12
44
18.5
15
6
14
44
18.5
16.5
6
14
44
19.5
15
6
16
44
19.5
16
6
16
44
19.5
16.5
6
16
44
21
15
6
16
44
21
16
8
16
44
21
16.5
8
16
44
21
17
8
16
46
19.5
15
6
16
46
19.5
16
6
16
46
19.5
16.5
6
16
49
21
16.5
6
20
49
21
17
6
20
49
21
20
6
20
In cases of lateral imbalance, you should follow the instructions here.
|
{
"pile_set_name": "Pile-CC"
}
|
How we view complainants; an ethical dilemma?
All too often, those patients who complain are thought to be unreasonable. Healthcare professionals often feel that patients do not have an understanding of the pressures and hardships that they struggle with on a day-to-day basis. When a patient complains, it is seen by the professional complained about as a wholly negative event, leading to loss of confidence and leaving that professional feeling demoralized. Often complaints are due to a breakdown in communication. Sometimes a patient is unhappy with a treatment charge or simply there is a perception that he/she has been poorly treated. The General Dental Council and Primary Care Trusts (and now their successors) take a dim view of dental practitioners who deal with complaints poorly. This article sets out to offer a different perspective on complaints, so that the complaint system can be used to build trust between dental professionals and patients, instead of instilling demoralization and fear of litigation into those on the receiving end of a complaint. This article is relevant to all dental professionals as complaints are an inevitability of practice.
|
{
"pile_set_name": "PubMed Abstracts"
}
|
Mayor Eric Garcetti ordered a halt Thursday to what he described as the Los Angeles Fire Department's "fatally flawed" recruitment process.
The move is part of Garcetti's ongoing reform effort and means the next scheduled LAFD class of 70 eligible cadets will expire. A new recruitment process won't begin until after a process review by the RAND Corporation.
2014 SoCal Images in the News
"I have determined that the fire department's recruiting process is fatally flawed," Garcetti said in a statement. "Reforming the fire department is a key part of my back to basics agenda, and the integrity of its recruiting process is vital to ensuring the department responds quickly, is technologically advanced, and reflects the city it serves in the future. Our firefighters perform heroically every day, and I want to make sure that the next generation of firefighters upholds the levels excellence practiced by today's firefighters."
The move disappointed members of United Firefighters of Los Angeles City, the union that represents firefighters.
Top News Photos of the Week
"It has been stated over and over again that restoration is the top priority for the fire department and staffing shortages should be the number one concern for our severely decimated department," said Frank Lima, the president of UFLAC, in a statement. "No new firefighters have been hired in over five years. A scarcity of personnel not only puts firefighter and public safety at risk, but it hurts the city's fiscal health as well."
Lima said firefighters are being forced to work overtime to make up for the staffing deficit. In the first two and a half months of 2014, 12 civilians have died in fire-related fatalities, he said. The city is on pace to set record levels of fire related deaths, Lima added.
"We have hit a tipping point where it just didn't make sense to continue to pay so much overtime without hiring new workers," he said. "It doesn't make financial sense and it's not safe."
Garcetti made the decision after it was revealed that department staff organized special recruiting workshops for "LAFD insiders," according to the statement from the mayor's office. At least two interview and resume preparation workshops were organized by a member of the LAFD, according to a statement from the department's interim Fire Chief James Featherstone.
"These workshops were advertised via LAFD email and were intended to be limited to members of the LAFD Cadet Program and family of LAFD members," Feathersone said. "While these actions may have been conceived in good faith, the result was a recruiting and hiring process that was less than fair and impartial."
The review of the recruitment process by Garcetti's office found classes with a "disproportionate amount of recruits" who were related to LAFD staff, some to senior managers who oversaw the recruitment and training process. Garcetti also cited a one-minute window allowed for filing paperwork that was part of the recruitment process.
The RAND Corporation -- a nonprofit institution that researches and analyzes policy making -- will make recommendations designed to reform the process. The LAFD also started an internal investigation.
|
{
"pile_set_name": "Pile-CC"
}
|
Quasi-null lens optical system for the fabrication of an oblate convex ellipsoidal mirror: application to the Wide Angle Camera of the Rosetta space mission.
The design of a quasi-null lens system for the fabrication of an aspheric oblate convex ellipsoidal mirror is presented. The Performance and tolerance of the system have been analyzed. The system has been applied successfully for the fabrication of the primary mirror of the Wide Angle Camera (WAC), the imaging system onboard the Rosetta, the European Space Agency cornerstone mission dedicated to the exploration of a comet. The WAC is based on an off-axis two-mirror configuration, in which the primary mirror is an oblate convex ellipsoid with a significant conic constant.
|
{
"pile_set_name": "PubMed Abstracts"
}
|
Aphonia and dysphagia after gastrectomy.
A 67-year-old male was referred to our otolaryngological clinic because of aphonia and dysphagia. His voice was breathy and he could not even swallow saliva following a total gastrectomy for gastric carcinoma performed 2 weeks previously. Laryngeal fiberscopy revealed major glottal incompetence when he tried to phonate. However, both vocal folds abducted over the full range during inhalation. The patient could not swallow saliva because of a huge glottal chink, even during phonation. Based on these findings, he was diagnosed as having bilateral incomplete cricoarytenoid dislocation after intubation. The patient underwent speech therapy; within 1 min his vocal fold movement recovered dramatically and he was able to phonate and swallow. There have been few case reports of bilateral cricoarytenoid dislocation, and no effective rehabilitation has been reported. We believe that our method of vocal rehabilitation serves as a useful reference for physicians and surgeons worldwide.
|
{
"pile_set_name": "PubMed Abstracts"
}
|
Q:
Hibernate multitenancy: change tenant in session
We're developing a SaaS solution for several consumers. This solution is based on Spring, Wicket and Hibernate. Our
database contains data from several customers. We've decided to model the database as follows:
public
Shared data between all customers, for example user accounts as we do not know which customer a user belongs to
customer_1
customer_2
...
To work with this setup we use a multi-tenancy setup with the following TenantIdentifierResolver:
public class TenantProviderImpl implements CurrentTenantIdentifierResolver {
private static final ThreadLocal<String> tenant = new ThreadLocal<>();
public static void setTenant(String tenant){
TenantProviderImpl.tenant.set(tenant);
}
@Override
public String resolveCurrentTenantIdentifier() {
return tenant.get();
}
@Override
public boolean validateExistingCurrentSessions() {
return false;
}
/**
* Initialize a tenant by storing the tenant identifier in both the HTTP session and the ThreadLocal
*
* @param String tenant Tenant identifier to be stored
*/
public static void initTenant(String tenant) {
HttpServletRequest req = ((ServletWebRequest) RequestCycle.get().getRequest()).getContainerRequest();
req.getSession().setAttribute("tenant", tenant);
TenantProviderImpl.setTenant(tenant);
}
}
The initTenant method is called by a servlet filter for every request. This filter is processed before a connection
is opened to the database.
We've also implemented a AbstractDataSourceBasedMultiTenantConnectionProviderImpl which is set as our
hibernate.multi_tenant_connection_provider. It issues a SET search_path query before every request. This works like charm for requests passing through the servlet filter described above.
And now for our real problem: We've got some entrypoints into our application which do not pass the servlet filter,
for instance some SOAP-endpoints. There are also timed jobs that are executed which do not pass the servlet filter.
This proves to be a problem.
The Job/Endpoint receives a value somehow which can be used to identify which customer should be associated with the
Job/Endpoint-request. This unique value is often mapped in our public database schema. Thus, we need to query the
database before we know which customer is associated. Spring therefore initializes a complete Hibernate session. This
session has our default tenant ID and is not mapped to a specific customer. However, after we've resolved the unique
value to a customer we want the session to change the tenant identifier. This seems to not be supported though, there
is no such thing as a HibernateSession.setTenantIdentifier(String) whereas there is a
SharedSessionContract.getTenantIdentifier().
We thought we had a solution in the following method:
org.hibernate.SessionFactory sessionFactory = getSessionFactory();
org.hibernate.Session session = null;
try
{
session = getSession();
if (session != null)
{
if(session.isDirty())
{
session.flush();
}
if(!session.getTransaction().wasCommitted())
{
session.getTransaction().commit();
}
session.disconnect();
session.close();
TransactionSynchronizationManager.unbindResource(sessionFactory);
}
}
catch (HibernateException e)
{
// NO-OP, apparently there was no session yet
}
TenantProviderImpl.setTenant(tenant);
session = sessionFactory.openSession();
TransactionSynchronizationManager.bindResource(sessionFactory, new SessionHolder(session));
return session;
This method however does not seem to work in the context of a Job/Endpoint and leads to HibernateException such as
"Session is closed!" or "Transaction not succesfully started".
We're a bit lost as we've been trying to find a solution for quite a while now. Is there something we've misunderstood?
Something we've misinterpreted? How can we fix the problem above?
Recap: HibernateSession-s not created by a user request but rather by a timed job or such do not pass our servlet
filter and thus have no associated tenant identifier before the Hibernate session is started. They have unique values
which we can translate to a tenant identifier by querying the database though. How can we tell an existing Hibernate
session to alter it's tenant identifier and thus issue a new SET search_path statement?
A:
We've never found a true solution for this problem, but chimmi linked to a Jira-ticket were others have requested such a feature: https://hibernate.atlassian.net/browse/HHH-9766
As per this ticket, the behavior we want is currently unsupported. We've found a workaround though, as the number of times we actually want to use this feature is limited it is feasible for us to run these operations in separate threads using the default java concurrency implementation.
By running the operation in a separate thread, a new session is created (as the session is threadbound). It is very important for us to set the tenant to a variable shared across threads. For this we have a static variable in the CurrentTenantIdentifierResolver.
For running an operation in a separate thread, we implement a Callable. These callables are implemented as Spring-beans with scope prototype so a new instance is created for each time it is requested (autowired). We've implemented our own abstract implementation of a Callable which finalizes the call()-method defined by the Callable interface, and the implementation starts a new HibernateSession. The code looks somewhat like this:
public abstract class OurCallable<TYPE> implements Callable<TYPE> {
private final String tenantId;
@Autowired
private SessionFactory sessionFactory;
// More fields here
public OurCallable(String tenantId) {
this.tenantId = tenantId;
}
@Override
public final TYPE call() throws Exception {
TenantProvider.setTenant(tenantId);
startSession();
try {
return callInternal();
} finally {
stopSession();
}
}
protected abstract TYPE callInternal();
private void startSession(){
// Implementation skipped for clarity
}
private void stopSession(){
// Implementation skipped for clarity
}
}
|
{
"pile_set_name": "StackExchange"
}
|
import java.util.*;
/**
* 因为离边界越远的节点越可能成为根节点，所以每次删除叶子节点，一层一层往里收缩，最后留下的就是可能的根节点
* 使用拓扑排序的类似算法，每次删除入度为1的节点，最后剩下的1个或两个入度为1的节点就是可能的最小高度树的根节点
*/
class Solution {
public List<Integer> findMinHeightTrees(int n, int[][] edges) {
List<Integer> res = new ArrayList<>();
//特殊情况：
if(n <= 2){
for(int i = 0;i < n;i++) res.add(i);
return res;
}
//构造邻接矩阵。注意，因为这里每次修改的节点都是叶子节点，只有一个邻接点，所以可以使用Set，如果有多个邻接点，则不能用Set
//因为Set不能边遍历边修改，会报ConcurrentModificationException
Set<Integer>[] graph = new Set[n];
//初始化
for(int i = 0;i < n;i++){
graph[i] = new HashSet<>();
}
//入度数组
int[] degrees = new int[n];
for (int[] edge : edges) {
graph[edge[0]].add(edge[1]);
graph[edge[1]].add(edge[0]);
degrees[edge[0]]++;
degrees[edge[1]]++;
}
//将入度为1的节点放入Queue
LinkedList<Integer> queue = new LinkedList<>();
for (int i = 0; i < n; i++) {
if (degrees[i] == 1)
queue.addLast(i);
}
//使得最后剩下一个或两个节点
while (n > 2) {
int size = queue.size();
//一次性减去所有的叶子节点
n -= size;
while (size-- > 0) {
//得到当前叶子节点
Integer cur = queue.poll();
//出度改变
degrees[cur]--;
//对于其所有的邻接点入度-1
if(!graph[cur].isEmpty()){
for(int node : graph[cur]){
degrees[node]--;
if(degrees[node] == 1) queue.add(node);
//删除叶子节点和邻接点的关系
graph[cur].remove(node);
graph[node].remove(cur);
}
}
}
}
//剩下位于Queue中的节点就是根节点
return queue;
}
}
public class $310_MinimumHeightTrees {
}
|
{
"pile_set_name": "Github"
}
|
Q:
Entity frame work code first relationship problems
Using the entity framework i am experiencing some problems with table relations.
It is between the two tables, Employee and AspNetUser.
I get the following error:
Unable to determine the principal end of an association between the types 'DB.EmployeeData' and 'DB.AspNetUser'. The principal end of this association must be explicitly configured using either the relationship fluent API or data annotations. My poco object can be seen below:
public partial class EmployeeData
{
public int Id { get; set; }
public string FullName { get; set; }
public string PostalCode { get; set; }
public string City { get; set; }
public double Longitude { get; set; }
public double Latitude { get; set; }
public int Age { get; set; }
public bool HighlightedNotificationsEnabled { get; set; }
public System.DateTime LastNotificationTime { get; set; }
public string StreetAndHouse { get; set; }
public string Country { get; set; }
public string Gender { get; set; }
public bool MatchingNotificationsEnabled { get; set; }
public bool NearbyNotificationsEnabled { get; set; }
public string VideoCVLink { get; set; }
public string FacebookUrl { get; set; }
public string InstagramUrl { get; set; }
public string LinkedInUrl { get; set; }
public string DriveUrl { get; set; }
public virtual AspNetUser AspNetUser { get; set; }
public virtual Translation Translation { get; set; }
}
public partial class AspNetUser
{
public AspNetUser()
{
this.AspNetUserClaims = new HashSet<AspNetUserClaim>();
this.AspNetUserLogins = new HashSet<AspNetUserLogin>();
this.AspNetRoles = new HashSet<AspNetRole>();
this.JobOffers = new HashSet<JobOffer>();
this.EmployeeQualifications = new HashSet<EmployeeQualification>();
this.NotifiedJobOffers = new HashSet<JobOffer>();
this.UserDevices = new HashSet<UserDevice>();
this.Applications = new HashSet<Application>();
}
public string Id { get; set; }
public string Email { get; set; }
public bool EmailConfirmed { get; set; }
public string PasswordHash { get; set; }
public string SecurityStamp { get; set; }
public string PhoneNumber { get; set; }
public bool PhoneNumberConfirmed { get; set; }
public bool TwoFactorEnabled { get; set; }
public Nullable<System.DateTime> LockoutEndDateUtc { get; set; }
public bool LockoutEnabled { get; set; }
public int AccessFailedCount { get; set; }
public string UserName { get; set; }
public string JobnetOrganisationId { get; set; }
public virtual ICollection<AspNetUserClaim> AspNetUserClaims { get; set; }
public virtual ICollection<AspNetUserLogin> AspNetUserLogins { get; set; }
public virtual ICollection<AspNetRole> AspNetRoles { get; set; }
public virtual EmployerData EmployerData { get; set; }
public virtual EmployeeData EmployeeData { get; set; }
public virtual ICollection<JobOffer> JobOffers { get; set; }
public virtual ICollection<EmployeeQualification> EmployeeQualifications { get; set; }
public virtual ICollection<JobOffer> NotifiedJobOffers { get; set; }
public virtual ICollection<UserDevice> UserDevices { get; set; }
public virtual ICollection<Application> Applications { get; set; }
}
I have googled this and tried to fix it using data annotations and fluent API but i cant get rid of the error.
Can anyone point out what i am doing wrong?
Thx :)
A:
It seems you have two issues. The first is that you have not defined a primary key on either table. So add the Key attribute to your Id properties. The second is that as you have a one-to-one relationship between the two objects you need to define the principal. In your case that would be:
modelBuilder.Entity<EmployeeData>()
.HasOptional(e => e.AspNetUser)
.WithRequired(a => a.EmployeeData);
|
{
"pile_set_name": "StackExchange"
}
|
Proinflammatory cytokine IL-1beta stimulates IL-8 synthesis in mast cells via a leukotriene B4 receptor 2-linked pathway, contributing to angiogenesis.
Recent studies have suggested that mast cells have critical roles in angiogenesis. However, the detailed mechanism by which mast cells contribute to angiogenesis is not yet clearly understood, especially in response to proinflammatory cytokines. In this study, we showed that the proinflammatory cytokine IL-1beta induces the synthesis of IL-8, a potent angiogenic factor, in human mast cells via the leukotriene B(4) receptor (BLT)2. We also characterized the BLT2 downstream signaling pathway and determined that BLT2-mediated IL-8 synthesis involves the upregulation of Nox1, a member of the NADPH oxidase family, Nox1-dependent reactive oxygen species generation and the subsequent activation of the redox-sensitive transcription factor NF-kappaB. For instance, knockdown of BLT2 and Nox1 with specific small interfering RNA, treatment with a specific BLT2 antagonist, LY255283, or treatment with a potential Nox inhibitor, diphenylene iodonium, suppressed IL-1beta-induced IL-8 synthesis. We found that the conditioned media collected from IL-1beta-treated human mast cell line HMC-1 had significantly enhanced angiogenic activity that could be dramatically attenuated by either small interfering RNA knockdown of BLT2 or treatment with neutralizing Ab to IL-8. Finally, the experiments were repeated using human primary cord blood-derived mast cells, and the results were clearly reproduced. Taken together, our results suggest that BLT2-Nox1-reactive oxygen species-dependent pathway plays a role in promoting the secretion of IL-8 from human mast cells in response to the proinflammatory cytokine IL-1beta, thus contributing to angiogenesis.
|
{
"pile_set_name": "PubMed Abstracts"
}
|
Abrasive blasting devices operate on the physical property that gas at a higher pressure flows towards and into gas at lower pressure. When abrasive powder is mixed with gas at higher pressure, the gas carries the abrasive powder as the gas accelerates and flows to the lower pressure. As the gas and abrasive powder blast the target material at high speed, the impact of the particles removes layers of the target material.
In dentistry this technology is known as micro-abrasion and is used to achieve a variety of goals—such as to remove foreign material or to dull a shiny surface, roughen or etch the surface to enhance bonding quality and to remove decay by drilling and cutting tooth structure.
Air abrasion devices date back decades with patented inventions by pioneers such as Ziegler U.S. Pat. No. 2,612,732, Crow U.S. Pat. No. 2,725,684, Schachter U.S. Pat. No. 3,626,841 and Paache U.S. Pat. No. 2,441,441.
Over the years two main approaches to air abrasion devices developed with Ziegler and Schechter following one approach and Crowe and Passche following another. One approach has been to provide a stationary mixing apparatus for generating the abrasive laden air stream and delivering the abrasive laden air stream through an extended hand-piece for directing the stream onto the target surface. Another approach has been to integrate the mixing apparatus into the hand-held device.
The first approach facilitates more complex mechanisms and many operational options since the size and weight of the device are of no concern. Because the extended hand-piece delivers the abrasive laden air stream independent of the mixing operation, the hand-piece can be held at any orientation during operation. Deardon et al. U.S. Pat. No. 6,083,001 discloses a dental air abrasion system in which the flow of the particles is electronically controlled by pressure differentials. Rainey U.S. Pat. No. 6,093,021 discloses an automated control system which utilizes a gas stream mounted particulate sensor to regulate fluid flow rates into and around the ultrasonically agitated mixing chamber in order to accurately maintain the abrasive concentration in the air stream. Various methods for reducing the overspray of the abrasive have also been developed for these devices. Ho U.S. Pat. No. 5,356,292, Coston U.S. Pat. No. 5,197,876, and Burns et al. U.S. Pat. No. 6,024,566 disclose add-on splatter guard and collector attachments to air abrasion devices.
In the second approach, the size, weight, and ergonomic shape of the device are significant factors. Herald et al. U.S. Pat. No. 5,199,299 and Burns et al. U.S. Pat. No. 6,439,966 disclose innovative hand-holdable air abrasion devices which mount the mixing apparatus into the hand-piece. The drawback of this approach is that the operation of these devices is limited by the orientation of the mixing chamber.
An adjunct to the second approach has been the concept of simple self-contained air abrasion devices—such as Hertz U.S. Pat. No. 5,839,946 (and its derivative U.S. Pat. No. 6,287,180, U.S. Pat. No. 6,951,505, and Granted Appl. No. 09/939,865), Groman U.S. Pat. No. 6,398,628 (and its derivative U.S. Pat. No. 6,347,984 and Pending application Ser. No. 10/144,228), Schur et al. U.S. Pat. No. 6,004,191, and Trafton et al. U.S. Pat. No. 6,354,924. These devices rely on the air stream to perturb the abrasive and generate the mixing action based on Stark et al. U.S. Pat. No. 4,475,370 fixed air abrasion device for treating dental castings.
Merging of Stark's blow-through mixing method into the hand-piece so the mixing chamber is held between the user's fingers has taken air abrasion art to a new level. Because of their simplified structures, simple self-contained air abrasion devices tend to be less expensive to manufacture and can therefore be marketed to the user as disposable instruments.
With increased emphasis in Medical, Pharmaceutical, Cosmetic and Dental applications on reduced cross-patient contamination, there has been a significant drive towards single usage disposable packaging and devices. With advances in materials and fabrication technologies the cost of these devices has been significantly reduced. Dougherty U.S. Pat. No. 4,391,590 discloses a syringe and stopper like cartridge device for dispensing material while Hertz U.S. Pat. No. 5,839,946 patent discloses the formulation an air abrasion instrument from a syringe and stopper type structure. Both innovations capitalize on the lower cost of fabrication and the well established production methods of a syringe and stopper configuration.
Simple self-contained prior art air abrasion devices support an elongated cylindrical chamber with an inlet conduit for delivering the air into the mixing chamber and a discharge conduit for carrying the air-abrasive mixture out of the mixing chamber. The mixing chambers are utilized as a reservoir for storing the abrasive powder. Once the reservoir is depleted of abrasive material, the devices are discarded and therefore function as disposable instruments which do not require sterilization post intra-oral use.
To prevent the abrasive material from escaping the mixing chamber or becoming contaminated prior to use, simple self-contained prior art air abrasion devices add additional components which seal the inlet and outlet ports and conduits. While the Hertz U.S. Pat. No. 5,839,946, and Schur et al. U.S. Pat. No. 6,004,191 devices include passive caps which must be removed prior to using the instrument, Hertz U.S. Pat. No. 6,951,505 and U.S. Pat. No. 6,287,180, and Groman U.S. Pat. No. 6,398,628 and U.S. Pat. No. 6,347,984 add functional components which actively prevent the abrasive from exiting the mixing chamber. Groman U.S. Pat. No. 6,398,628 has a filter that prevents the abrasive from exiting the device's inlet port and a movable discharge conduit which prevents abrasive material from exiting the mixing chamber when the discharge conduit inlet port abuts the side wall of the mixing chamber. Groman pending application Ser. No. 10/144,228 support a deformable gasket at the discharge port internal to the mixing chamber which opens when flow is present. Hertz U.S. Pat. No. 6,951,505 has a deformable seal at the inlet port external to the mixing chamber which functions as a check-valve that allows the pressurized-gas to enter the instrument but prevents abrasive from exiting the instrument. Groman U.S. Pat. No. 6,398,628 discloses a deformable and movable cap configurations which block both the delivery conduit inlet and discharge conduit outlet prior to use.
Another disposable delivery method disclosed by Zhang et al. U.S. Pat. No. 6,343,717 attempts to address the containment of stored material utilizing a pipette structure. A typical pipette consists of a slender pipe or tube that is used to transfer or measure small quantities of material from one location to another. The most common type of pipette consists of a small tube that widens into a bulb at the middle.
Zhang et al. pipette structure is made of a rigid or resilient material that is pre-filled with a pharmaceutical or cosmetic product and is used once and then discarded. Zhang et al. discloses a plurality of ways by which the disposable pipette can be sealed to contain the material and then unsealed by the user prior to use for dispensing the stored material. According to Zhang's et al. invention the majority of material is retained within the bulb section of the pipette, but Zhang's et al. sealing methods permit the contained material to migrate into the top and bottom tube sections. Although Zhang's et al. use of a pipette structure leads to a very cost effective means of delivering the contained material, Zhang's et al. sealing methods are not compatible with the needs of air abrasion devices.
Pressurized air stream is delivered to the simple self-contained air abrasion devices of Hertz, Groman, Schur, and Trafton via custom connectors which engage the device externally and to form a seal with the device body to deliver the pressurized air to the mixing chamber delivery port. The connectors are designed to supply clean dry air in order to maintain the abrasive powder dry, since any moisture causes clumping of the abrasive material and therefore the malfunction of the device. The dry air is required because the gas delivery conduit leads directly into the mixing chamber; therefore any liquid present at the entry to the device gets trapped in the mixing chamber. Hertz et al. U.S. Pat. No. 6,293,856 discloses a connector with additional conduits for carrying other types of fluids passively through the mixing chamber. This configuration requires a very complex connector to assure the separation of the fluids delivered to the air abrasion instrument without contaminating the mixing chamber. Custom connectors which supply dry air add to the installation cost and complexity of these disposable devices. And because they attach to the body of the devices, these connectors are typically very bulky.
Referring to FIG. 1, prior art self-contained air abrasion devices use a blow-through methodology to agitate the abrasive powder. More specifically, these devices utilize the delivery conduit to deliver the gas stream into the abrasive material. As the gas stream blows through the abrasive material, the abrasive material is agitated. Gravity is utilized to assure that the non-aerated abrasive remains at the bottom of the mixing chamber. As the air stream reverses direction towards the discharge conduit inlet, aerated particles are captured by the air stream. The abrasive laden air stream is pushed out of the mixing chamber through the discharge conduit by the higher pressure gas source.
In their reduction to practice, both the Schur and Groman devices require the user to maintain the orientation of the device so the mixing chamber points downward. The attached user instructions for the Schur and Groman devices outline the specific user instructions cautioning the user about mis-orienting the mixing chamber. To compensate for his shortcoming, the marketed Groman instrument provides a finger bendable discharge conduit. The marketed Schur device provides a bending tool, so the user is able to form the delivery conduit to reach upper surfaces while maintaining the proper orientation of the mixing chamber.
Referring to FIG. 2, if the user attempts to utilize these prior art devices with the mixing chamber horizontal or upside down, the abrasive material is pushed directly into the discharge conduit without being properly mixed with the air steam. This leads to a concentration of abrasive material to exit the device in an uncontrolled manner, which creates a cloud of abrasive dust or clogs up the discharge conduit as the abrasive powder binds. Additionally, in certain orientations the delivery conduit is not immersed in the abrasive material which also disrupts the mixing operation of these prior art devices. In fact, the pressurized-gas exiting the delivery conduit creates a back pressure on the abrasive within the mixing chamber causing the abrasive powder particles to bind together instead of mix with the air stream. Most importantly, these disruption in flow can lead to a defective clinical procedure which either under or over etches the target tooth surface.
|
{
"pile_set_name": "USPTO Backgrounds"
}
|
Q:
Which type of adverb phrase is it? Could you give me some similar explanatory examples please?
A fascinating world of scientific wonders, the amphibian species is
full of unusual and extreme adaptations and is home to numerous
unsolved mysteries.
Amphibians
A:
A fascinating world of scientific wonders, the amphibian species is full of unusual and extreme adaptations and is home to numerous unsolved mysteries.
I'm not at all sure about the factual content of your example. No matter, it works like this:
It's not an adverb phrase. The sequence in italics is a noun phrase in predicative adjunct function: predicative because it is related to the predicand, "the amphibian species", and adjunct because it is an optional item in clause structure.
Predicative adjuncts are not restricted to noun phrases: they can also be adjective or preposition phrases:
[1] Unwilling to accept the terms, Ed resigned.
[2] In a bad temper, as usual, John walked on ahead of the main party.
The italicised sequence in [1] is an adjective phrase that relates to a predicand (the subject, “Ed”). Likewise in [2], the italicised preposition phrase relates to a predicand (the subject, John). The adjuncts ascribe some property to Ed and John respectively.
|
{
"pile_set_name": "StackExchange"
}
|
Q:
Repeated single or multiple tests with Nose
Similar to this question, I'd like to have Nose run a test (or all tests) n times -- but not in parallel.
I have a few hundred tests in a project; some are some simple unit tests. Others are integration tests w/ some degree of concurrency. Frequently when debugging tests I want to "hit" a test harder; a bash loop works, but makes for a lot of cluttered output -- no more nice single "." for each passing test. Having the ability to beat on the selected tests for some number of trials seems like a natural thing to ask Nose to do, but I haven't found it anywhere in the docs.
What's the simplest way to get Nose to do this (other than a bash loop)?
A:
You can write a nose test as a generator, and nose will then run each function
yielded:
def check_something(arg):
# some test ...
def test_something():
for arg in some_sequence:
yield (check_something, arg)
Using nose-testconfig, you could make the number of test runs a command line argument:
from testconfig import config
# ...
def test_something():
for n in range(int(config.get("runs", 1))):
yield (check_something, arg)
Which you'd call from the command line with e.g.
$ nosetests --tc=runs:5
... for more than one run.
Alternatively (but also using nose-testconfig), you could write a decorator:
from functools import wraps
from testconfig import config
def multi(fn):
@wraps(fn)
def wrapper():
for n in range(int(config.get("runs", 1))):
fn()
return wrapper
@multi
def test_something():
# some test ...
And then, if you want to divide your tests into different groups, each with their own command line argument for the number of runs:
from functools import wraps
from testconfig import config
def multi(cmd_line_arg):
def wrap(fn):
@wraps(fn)
def wrapper():
for n in range(int(config.get(cmd_line_arg, 1))):
fn()
return wrapper
return wrap
@multi("foo")
def test_something():
# some test ...
@multi("bar")
def test_something_else():
# some test ...
Which you can call like this:
$ nosetests --tc=foo:3 --tc=bar:7
A:
You'll have to write a script to do this, but you can repeat the test names on the commandline X times.
nosetests testname testname testname testname testname testname testname
etc.
A:
Solution I ended up using is create sh script run_test.sh:
var=0
while $1; do
((var++))
echo "*** RETRY $var"
done
Usage:
./run_test.sh "nosetests TestName"
It runs test infinitely but stops on first error.
|
{
"pile_set_name": "StackExchange"
}
|
1. Field of Invention
This invention relates to vial closures, specifically to such closures which are used for vials containing medicinal pills.
2. Description of Prior Art
Pharmacies commonly dispense prescription medications in the form of pills. These pills are typically packaged in vials; that is, containers that are sealed with a closure. Two common types of closures are the simple snap-fit type, which is not resistant to opening by a child, and the “child-resistant” type, of which there are several different designs in use.
Ordinarily, prescription medication must be taken at more-or-less regular intervals. Failure to do so can result in ineffective treatment or other serious consequences, such as an overdose. Consequently, it is important for patients to not forget the number of pills they have taken.
It is well known in the art to incorporate some type of indicator into
In recent years there has been a number of patents granted for electronic timers with alarms that have been incorporated into closures. These devices all suffer from higher cost and greater complexity than simple mechanical solutions.
Mechanical devices in the prior art that incorporate indicators into closures invariably indicate time. The indicator in such devices would be set to the next time to take a pill, for example. U.S. Pat. No. 5,279,422 to Adams this type of indicator. As it turns out, the vast majority of prescriptions for pills are written for 2, 3, 4, or 6 pills per day to be taken. Thus, it is really only necessary for a person to be reminded of how many pills they have already taken that day; this is far more useful than a time indication of when to take the next pill, or when the last pill was taken, especially because most people are probably not extremely precise about the time when a pill is taken.
Moreover, the devices shown in the prior art lack the necessary simplicity and low cost to be incorporated into the inexpensive packaging that is used for common prescriptions. Or, they lack the ability to be easily adapted for both the simple snap-fit type and child-resistant type of caps.
|
{
"pile_set_name": "USPTO Backgrounds"
}
|
Michelle Obama was recently asked to the Marine Corps Ball, proving that Marines are indeed lusting after someone other than Mila Kunis and Justin Timberlake, and she said yes!
And judging from the above photo, it's fairly easy to see why (sure, maybe her reasons were different but hey, she has eyes!) Via Politico:
20 year old Marine Lance Corporal Aaron Leeks from Frederick, Maryland has asked First Lady Michelle Obama to accompany him to the Marine Corps Ball next November. "With your husband's permission of course," said Leeks. The first lady responded, "I'd love to" and brought an aide over to get his information. The first lady met Leeks at a Toys-for-Tots event at Joint Base Anacostia-Bolling located in southeast Washington, DC.
Okay, I'm all for fidelity but goodness if that isn't the type of super polite "Can I borrow you from your husband proposal?" that would make a girl non-cheating-swoon for at least 5 minutes.
Press Secretary Jay Carney said he couldn't confirm that Mrs. Obama could attend but that, "The first lady's commitment to military families is very strong indeed. So I'm sure she was flattered by the invitation."
G/O Media may get a commission Subscribe and Get Your First Bag Free Promo Code AtlasCoffeeDay20
A commitment that's probably even easier to keep when the invitation comes from a polite would-be leading man in a movie I'm currently writing in my head.
Michelle Obama asked to Marine Corps Ball [Politico]
|
{
"pile_set_name": "OpenWebText2"
}
|
Com cinco episódios, ela vai mostrar violão para tocar com um braço só, fogão solar e até um pão feito com farinha de barata.
|
{
"pile_set_name": "OpenWebText2"
}
|
James Events Receives Praise as the Best Corporate Event Planner
Press Release -
May 22, 2014
James Events, the leading event planners in Orange County, have been praised for their exemplary corporate planning skills. The company has been organising occasions for leading companies for over twenty-five years.
James Events, the leading event planners in Orange County, have been praised for their exemplary planning skills for corporate occasions. The company has been organising parties for leading companies for over twenty-five years and has built a reputation for delivering the best most successful conferences and parties in the area.Corporate fund raisers, promotional days, staff motivational gatherings and company picnics and parties are all delivered with a touch of flair, innovation and class, thanks to the fantastic team. The praise the company has been receiving is increasing interest in its services and in recent months it has seen an increase in enquiries and bookings.James Events is dedicated to giving clients the party that surpasses their expectations. Every detail is taken care of in a professional manner and nothing is left to chance. Large vents can be highly stressful to organise and pull-off, but thanks to the company, clients are enjoying stress free hosting of some incredibly large and important parties. Amongst the corporate clients, there are requests for launches, charity balls, company picnics and conferences are the types of occasionthe company will deliver.
The most popular aspect of the company, and one of the main reasons they are chosen by leading companies and organisations, is that they take care of everything. They are a single source service. This means that all the suppliers, for food, entertainment, services, venues, and so on, are all sourced, booked and managed by the company. The company has a huge network of highly gifted and recommend suppliers and service providers which they use to make their occasions stand out from all the rest. It is no surprise the company is receiving such praise from its customers.A spokesperson for James Events said, "We're delighted by the positive reviews and praise we are receiving! We want to thank all our clients for their wonderful feedback. We've been in this business for over twenty-five years and are proud of our reputation as leading party organisers in Orange County. It is our mission to make each occasion unique, memorable and fun for everyone concerned! For more information on our services, please go to your website where you can read about what we do and how we can help you throw the party of the century!"
About us:James Events has the knowledge and experience to make your vision come to life with precision and detail. Our talented team of event experts handles every detail: location, food, entertainment, logistics and little things that can make your event special. There are truly no limits to what a James Event can be. For professional conferences, luncheons, team-building events, award dinners, facility openings, sales meetings and holiday parties, work with James Events so you can have successful events, every time. For more information please go to http://jamesevents.com/
Categories:
Tags:
Press Contact
About James Event Productions
Over our 25-year history, we have produced more than 14,000 events for leading companies, non-profits and schools including The Walt Disney Companies, In-N-Out Burger, Auto Club of Southern California, Kaiser Permanente and Nickelodeon. James Events has the knowledge and experience to make your vision come to life with precision and detail. Our talented team of event experts handles every detail: location, food, entertainment, logistics and little things that can make your event special.
|
{
"pile_set_name": "Pile-CC"
}
|
Q:
How do I get the Last column of the array to add up at the end?
The part of the code that is getting me stuck is the end print out and getting it to add the BILL parts of the array for the final system.out message and display it correctly. The array and the rest of the program work fine it is just that last part that I am having trouble with.
private static void printBilling() {
int numberOfCustomers = Accounts.length;
int sum=0;
System.out.println("\n\nTable showing charges for cell phones");
for (int customer=0; customer<numberOfCustomers; customer++){
System.out.printf ("%s %s %s:\n GB used = %s Please Pay %s \n\n",
Accounts[customer][NAME],
Accounts[customer][ACCT],
Accounts[customer][SELECTION],
Accounts[customer][USED],
Accounts[customer][BILL]);
}
sum+= Accounts.length[customer][BILL];
{
System.out.printf ("The total owed is %s", +sum);
}
A:
your bill probably is a String then change like below.
if it contains integer, use this
sum+= Integer.parseInt(Accounts[customer][BILL]);
if it contains doubles, need (double sum variable) then use
sum+= Double.parseDouble(Accounts[customer][BILL]);
|
{
"pile_set_name": "StackExchange"
}
|
Q:
SOAPHandler handleFault
Can any one explain me when SOAPHandler's handleFault(SOAPMessageContext context) method is called ?
My Handler class is :
public class WebServiceHandler implements SOAPHandler<SOAPMessageContext> {
private void dumpSOAPMessage(SOAPMessage msg) {
if (msg == null) {
System.out.println("SOAP Message is null");
return;
}
System.out.println("");
System.out.println("--------------------");
System.out.println("DUMP OF SOAP MESSAGE");
System.out.println("--------------------");
try {
ByteArrayOutputStream baos = new ByteArrayOutputStream();
msg.writeTo(baos);
System.out.println(baos.toString(getMessageEncoding(msg)));
} catch (Exception e) {
e.printStackTrace();
}
}
@Override
public boolean handleMessage(SOAPMessageContext context) {
try {
dumpSOAPMessage(((SOAPMessageContext) context).getMessage());
} catch (Exception e) {
e.printStackTrace();
}
return true;
}
@Override
public boolean handleFault(SOAPMessageContext context) {
// TODO Auto-generated method stub
System.out.println("Inside handle fault:: " + context);
return true;
}
@Override
public void close(MessageContext context) {
// TODO Auto-generated method stub
}
@Override
public Set<QName> getHeaders() {
Set<QName> set = new HashSet<QName>();
return set;
}
}
Can any one explain when handleMessage and handleFault method is called ?
A:
According to : http://docs.oracle.com/cd/E13222_01/wls/docs103/webserv_adv/handlers.html#wp222524
handleMessage :
The Handler.handleMessage() method is called to intercept a SOAP
message request before and after it is processed by the back-end
component
and
handleFault :
Implement this method to handle processing of any SOAP faults
generated by the handleMessage() method, as well as faults generated
by the back-end component.
|
{
"pile_set_name": "StackExchange"
}
|
Jewish Singing Harmonies In Prayer
Ralph Northam (D) joined Jewish congregants and clergy of many faiths in singing Jewish words of sorrow and resolve. Hogan.
Wanna sing? Our Choir Auditions page is the place to help you find the perfect outlet for your singing ambitions. Ads here are for groups listed in the VAN Choir Directory. Check the Info Exchange for ads from ensembles not listed in the Choir Directory and for ads for professional singers.
The two-hour service—which was filled with songs and prayers for peace and harmony—included religious leaders from. In a litany for love and justice, Muslims, Christian and Jewish leaders encourage.
The Psalms Sung: The Power of Music in Sacred Worship; The Psalms Sung: The Power of Music in Sacred Worship. Front Matter. The Power of Music in Sacred Worship. the revelations of Joseph Smith have connected prayers with singing, reminiscent of the great Psalm prayers of the Old Testament: “Yea, the song of the righteous is a prayer.
Reading today about congregants hiding in closets and offices and classrooms where I spent formative years learning, singing.
Hebrew in Harmony features music from today’s top Jewish musicians. Along with arts, movement, videos, and a full digital learning experience, this multimedia curriculum invites students to learn to sing and pray in Hebrew, as well as read, explore, and interpret prayer text.
I am heartened by a flood of calls from allies of all faiths, who stand with the Jewish community at this critical time. T.
Dozens of people stood in solidarity with the Jewish community and those affected by the shooting Saturday night in Union Squ.
The Chichester hit factory triumphs again. The country’s most consistently classy regional theatre has gained a well-deserved reputation in recent years for its top quality revivals of musicals and, under new artistic director Daniel Evans, it is tuneful business as usual.
The official site of the life and legacy of Debbie Friedman (1951 – 2011). This site is dedicated to Debbie’s beloved Jewish music, accomplished career, and.
To put your name or the name of another person on the Rebbe’s Daily Healing Prayer list, please click on the CONTACT ME link and give the FIRST NAME of the person needing healing prayers and the FIRST NAME (if known) of the persons MOTHER (we Jews say healing prayers in the merit of the mothers).
Their unfathomable pain is shared by the entire Jewish. prayers lead by Rabbi Mendy Weiss and Yzippy Weiss, and a message.
Four cantors sing songs of Passover in the "Pesach Extravaganza. Manilow wrote a musical about a Jewish musical group in Nazi Germany titled "Harmony" and, in the musical, wrote the song "Where You.
Live Music And Dinner Minneapolis Live Music Venue Franchise School of Rock boasts more than 25,000 students worldwide, thanks to our concerted efforts to grow our music education franchise system to the next level. School of Rock boasts more than 25,000 students worldwide, thanks to our concerted
A. Transition of Prayer form Temple worship to synagogue to Church: 1. Temple worship directed the Jews to physically face Jerusalem in their personal prayers. a.
Zombie Movie Disney Musical Just within the next two weeks, we’ll see the return of the biggest basic cable hit in history — The Walking Dead, shambling back for its midseason premiere — and a new zombie movie starring Ellen Pag. Live Music Venue Franchise School
The fact that the shooting happened on a Saturday, or Shabbat, the Jewish sabbath, was especially significant to many. Observ.
Bend the Arc: Jewish Action, a progressive advocacy group, visited vigils from Boston to New Jersey to Minneapolis. In Squirr.
The crowd of more than 100 sang songs and recited Jewish prayers. The event was organized by progressive. click to enlarge.
The BESY Choir Gives Incredible Rendition Of ‘The Prayer’! | “I will instruct you and teach you in the way you should go; I will counsel you with my loving eye… “I will instruct you and teach you in the way you should go; I will counsel you with my loving eye on you.”(Psa 32:8) A choir of gospel singers provides soulful harmonies to.
For the second time in three days, people gathered in Vallejo on Tuesday to stand in solidarity with the Jewish community in.
On Valentine’s day, 3,000 Muslims, Jews and Christians gathered together in Haifa, Israel to sing in harmony in a bid to strengthen the hope for the ever-elusive peace in the Middle East. The participants have never met before but willingly signed up to learn to sing “One Day” by Matisyahu in English, Hebrew and Arabic. The result is a visual and visceral experience that is nothing short of beautiful.
This group of castoffs formed a prayer circle outside Beth Israel’s westernmost driveway gate. For at least an hour, the stra.
I leave the consequences to your discretion, Heavenly Father. I will sing in harmony with you. In Jesus Name and with His help. Amen. “I am praying not only for these disciples but also for all who wi.
Encyclopedia of Jewish and Israeli history, politics and culture, with biographies, statistics, articles and documents on topics from anti-Semitism to Zionism.
The best and largest collection of commentaries on the internet, including every commentary that Charles Spurgeon recommended. See more About this Collection.. Whole Bible Commentaries Whole Bible Commentaries 60+ Bible Background, Survey & Introduction 78+ Biblical Theology 30+. Old Testament Commentaries 1,200+ Commentaries on the Whole Old Testament 12+..
Located west of the UO campus, Chabad serves as a religious and community meeting space for Jewish UO students and community members. Gurevitch was among more than 100 people to attend a community pra.
Books for PJ Library are chosen from among the very best of Jewish children’s books. A committee of writers, early childhood professionals, and Jewish content specialists seek out compelling stories, captivating illustrations, and diverse perspectives on Jewish customs and practice.
Spanish and Portuguese Jews, also called Western Sephardim, are a distinctive sub-group of Iberian Jews who are largely descended from Jews who lived as New Christians in the Iberian Peninsula during the immediate generations following the forced expulsion of unconverted Jews from Spain in 1492 and from Portugal in 1497. Although the 1492 and 1497 expulsions of unconverted Jews from Spain and.
Historical Blues Venues In Chicago Mississippi Delta blues, also known as Delta blues, regional style of early 20th-century American folk music, centred in the Delta region of northwestern Mississippi.The pioneers of the style played a key role in developing the market for traditional blues recordings in the
Indie-pop band Lola Marsh has the music world delighting in its perfectly synced harmonies. The band formed in Tel Aviv in 2013 as a duo (Gil Landau/guitars, keyboards and Yael Shoshana Cohen/vocals), and today is comprised of five-members.
They were invited to lead the massive assemblage, first with a prayer from Psalm 20 and then the singing of Ani Ma’amin, a Hebrew song affirming the Jewish belief in the coming of the Messiah.
Among them were Jewish Clevelanders offering support and seeking solace. warmth of her congregation and promised it would.
Dance Schools West Chester Pa The Pointe Dance Studio, The Christina Cultural Arts Center, DCNS Summer Intensive, Booker T. Washington High School, Muhlenberg and Dean College and West Chester and Temple University. She has worked. Others use local fruit, fresh herbs, spices, nuts, coffee, tea, or other
From 1995 to 2015 the Jewish Theatre in Stockholm was an innovative stage for the exploration of drama, dance, film, music and performance merging different art forms with technology and architecture. This is the archive.
So began the litany of prayers offered by grieving students. and with about 1 percent of its more than 4,000 students identifying as Jewish. The school’s campus ministry staff also includes a Musli.
But there is harmony, too, and we have the chance to savor and support it. HaZamir, the International Jewish Teen Choir. When they, together, sing David Burger’s iconic “Tefilah,” the prayer on beh.
PITTSBURGH (Reuters) – Some 2,500 people gathered on Sunday at a memorial service for the 11 Jewish worshipers slain in their.
Did John Calvin discourage harmonies in congregational singing? He did not discourage congressional singing, but it seems that John Calvin did discourage congressional singing in harmony. In the area of church music, Zwingli, although an amateur musician himself, had not set down any of his ideas of congregational singing.
I concluded part 1 by promising to share what I believe to be the most undervalued prayer. 34 Jewish men and 17 Jewish wom.
Singing with a group of friends is not only a fun, creative and rewarding way to spend some time but more and more studies are showing the various health, social and general well being of group harmony singing.
North Carolina Schools For Composing Music Opportunities for musicians can often feel few and far between, but there are actually a lot of options available to those with a Music Major. Most people who hold a music degree look towards either teaching or performing music, but there are
|
{
"pile_set_name": "Pile-CC"
}
|
FEATURED AUTHOR
ASSASSIN SHIFTER BOOK 1-6 BOXED SET REVIEW!!
MIKE AND I ARE HUGE FANS OF THIS SERIES, AND I AM SO STOKED THAT NEW FANS WILL BE INTRODUCED TO THIS SERIES VIA BOXED SET! WHAT A GREAT IDEA! HERE IS MY REVIEW OF THE FIRST BOXED SET, :) LET'S GET STARTED:
I THINK THIS QUOTE APPLIES TO ALL THESE STORIES! LET'S LOOK AT THE REVIEW! The Assassin Shifter Books 1-6 by: Sandrine Gasq Dion reviewed by: BeckyMateo Esposito loves his job. Hired assassin for the U.S. Government, he takes lives and he kicks ass with no mercy.When a job lands in his lap that's just not quite right, Mateo finds himself questioning orders for the first time in his career. Who would have known he'd be undone by a purple hippo...
Riley Flynn is CEO of Flynn Electronics. Deep in his closet, Riley wonders what it would feel like to be with a man. When his path crosses with Mateo's, their lives will never be the same. Because, Riley is Mateo's next target.... ~ Alaska With Love Josh Montgomery is a red-blooded American male that loves his job and loves women even more. Part of an elite team run by General Derek Jacobs, Josh and his fellow assassins take pride in their jobs. When one Assassin gig turns into a rescue mission on the side, Josh's life is turned upside down by the man he rescues. Now, he has to take a good long look at who he really is. With the help from his friends Mateo, Riley and Troy, Josh tries to understand what he's feeling. Mark Patterson joined Doctor's without Borders to do good in a foreign country. Little did he know a maniacal drug lord would kidnap him and use him to deal drugs, among other things.... Without any hope of rescue, Mark resigns himself to his prison. Until early one morning when he's rescued by a blonde God. Josh Montgomery is tall, light and beautiful....and straight. Being in close quarters only makes things harder on both men, will Mark get his man? Will Josh take a chance.... In the wilds of Denali Forest the two men come together, but is it for good? Or will Mark's past drive them apart? BY THE LIGHT OF THE MOON Dakota Cadotte is on the run and alone. Forced out from his pack by the Alpha for being gay, Dakota fends for himself in the Denali wilderness. When he ventures too far away from his safe zone, he's put right in the path of a truck on the highway and right into the arms of Sam Waters. The two hit it off immediately as man and wolf, but Dakota fears Sam won't be so understanding when he sees what he truly is. When Dakota shifts in front of Sam, the sparks fly, but Dakota's past and pack threaten their new-found relationship. HALF MOON RISING They say you can never go home... After more than a decade, Troy Bishop is going home to the family he left behind when his parents died. Little does Troy know, he's about to get a lot more than he bargained for.... Sawyer Quinton has been waiting for his mate for as long as he can remember, and was beginning to think he'd never find him. Until Troy Bishop walked back onto the Reservation. The attraction is immediate for both men, but one is hiding a secret and the other must finally open is heart. Sometimes you can't go back home....and sometimes when you do...you never leave. BEST LAID PLANS The men of the super-secret Skull Blaster unit have put down their weapons long enough to find love - and even husbands - from Arizona to Alaska and the Olympic Peninsula. Now their new families and friends are hoping to find their own happily-ever-afters. Werewolf Sawyer Quinton’s brothers have finally found their mates in the Drake brothers, Dakota’s former packmates. Things get off to a rocky start but love conquers all. Or does it? Can the Drakes overcome the gay bashing they suffered at their father’s hands and be true to themselves and the men they love? Or are some lessons just too hard to unlearn? FOR THE LOVE OF CADEN Caden Fournier has everything: Money, fame and an international business that's skyrocketing. But the one thing he really wants, money can't buy: Love. When fate puts him in the path of a speeding semi, Caden meets handsome paramedic Kellan Brady. Kellan Brady has been waiting for his mate for as long as he can remember. Always wanting the fairy tale of true love, he ends up instead with an injured and angry Frenchman pinned in his BMW.
If they can get past their rocky start, can the two men build a relations
Oh how much I love this series! One of the things I love about it is how well the stories fit together and flow into each other! Book 1, A Marked Man, is about Riley and Mateo, and it is a perfect introduction to this series, you will fall in love with sweet Riley, and sexy Mateo will have you all in a tizzy just like he does Riley! Can't wait for you guys to read this one. Book 2 Alaska With Love: This book will tug at your heartstrings right off the bat! Josh rescues Mark from captivity and doesn't understand the feelings he has for a man, and Mark have endured unspeakable cruelty, isn't looking for love for sure, but isn't that when love finds you, when you aren't looking? This one will hit you in the heart for sure! *5 Stars*Book 3 By the Light of the Moon: This is my favorite book in the series, Dakota has been forced out of his pack for being gay, Sam hits Dakota in wolf form, and takes care of him. Little does he know that the sexy guy he meets at a bar, and the wolf he has gotten attached to are one in the same! I fell in love with Dakota, and Sam takes such good care of him! *5 Stars* Book 4 Half Moon Rising: Troy goes back home after a long absence and finds Sawyer and gets more than he ever thought he wanted, but it turns out to be just what he needs, can Troy handle when Sawyers secret comes out? This one has it all, suspense, romance and is such a hot book! *5 Stars* Book 5 Best Laid Plans: Man oh man is this two stories for the price of one! These two sets of brothers are mates, but they couldn't have grown up more differently! The Quinton brothers go to meet their mates and they show up at the wrong time! Can the Drake brothers redeem themselves in their mates eyes? *5 Stars* Book 6 For the Love of Kaden: This one is great, though this couple gets a rocky start, you should never count love out! Kellan has finally found his mate, but has he found him to late? *5 Stars*So as you can guess my overall rating of books 1-6 is *5 Stars* or should I say:
TAKE A LOOK AT WHAT'S NEW FROM LANE HAYES!
THE SPENCER COHEN SERIES
OUT NOW!!
BEATEN TRACK PUBLISHING: FEATURED PUBLISHER
FEATURED AUTHOR
About Us
Hey friends, I'm Bec, and I am so happy to be bringing you reviews on the best mm books of the genre, I know you will find both authors new to you, and ones that are your faves, so join me and let's have a blast!!
Gadget
This content is not yet available over encrypted connections.
Google+ Followers
DISCLAIMER
Per the FTC regulations, please note that Bike Book Reviews receives books for review for free from publishers, authors, and publicists. This blog does not receive monetary compensation for reviews, and all reviews state honest opinions.
|
{
"pile_set_name": "Pile-CC"
}
|
The overall goals of this project are to determine whether (a) a special orthophosphate preparation (slow-release potassium phosphate or UroPhos-K) would produce physiological-physicochemical correction, inhibit stone formation, and provide safety of usage in absorptive hypercalciuria, and (b) a novel citrate formulation (potassium-magnesium citrate or K-Mag) would inhibit stone formation and provide safety of usage in patients with calcium stones. In Aim 1, we shall examine the effect of UroPhos-K on the crystallization of stone-forming salts while patients are kept on a constant metabolic diet. Aim 2, to commense upon satisfactory completion of Aim 1, will be a long-term randomized trial between UroPhos-K and placebo on stone formation. In Aim 3, we shall continue ongoing randomized trial between K-Mag and placebo on stone formation. Finally, Aim 4 will assess the value of K-Mag in preventing hypokalemia and magnesium depletion in patients taking thiazide for hypercalciuric nephrolithiasis. This project is biomedically important, since it could lead to the development of UroPhos-K as a rational and effective treatment of absorptive hypercalciuria, and of K-Mag as a superior drug over potassium citrate.
|
{
"pile_set_name": "NIH ExPorter"
}
|
Introduction
============
Background
----------
To increase the volume of the lips, a procedure called lip augmentation, or augmentation cheiloplasty, can be performed surgically or by way of a minimally invasive injection of filler material. Increasing the volume, restoring a symmetric ratio of all four lip quadrants, and reducing the fine lines and dryness of the lips are the primary indications for aesthetic cheiloplasty. Indications for augmentation cheiloplasty are to restore facial symmetry, to increase lip volume and fullness, and to improve definition of the vermilion border. Reasons that patients present for augmentation cheiloplasty are either medical (asymmetric face after trauma, nerve palsy, or cancer) or aesthetic: patients may want to improve fullness (and may already have a good shape of the lips), enhance atrophy (due to aging or genetics), or increase definition of the vermilion border (associated with smoking, a habit of inverting the lips in stress, and age).[@b1-ccid-7-191]
Augmenting the perioral region with injectable fillers is most often performed based on experience and by palpation, without knowledge of the exact, final position of the injected material in vivo. Not all injecting physicians are aware of the crucial subcutaneous structures.
To our knowledge, there is only one in vivo investigation on the exact positioning of injectable fillers,[@b2-ccid-7-191] and reports of severe complications, such as lip hematoma or thromboembolism, are rarely published.[@b3-ccid-7-191],[@b4-ccid-7-191] The lips, and especially the modiolar region, have strong blood supply[@b5-ccid-7-191] and are thus especially prone to hematoma and injury.[@b6-ccid-7-191]
Histological and imaging studies of lip and perioral anatomy
------------------------------------------------------------
There is a border line between the perioral epithelium and the red lip vermilion, which, on the upper lip, is called Cupid's bow and always has an individual, characteristic pattern. Portrait artists know this line as "the finger print of a face", which makes every smile unique. The upper lip shows a fine groove, the philtrum. Scar formations change the anatomy and tissue architecture[@b7-ccid-7-191] and have to be considered in lip augmentation because of altered distribution of the filler material.
The histology of the perioral region -- as in every type of skin (stratified squamous cell epithelium) -- consists of epidermis, dermis, and hypodermis. It contains hair follicles as well as sebaceous and sweat glands and is separated from the underlying tissue by connective tissue and muscles. On the dental side of the lip, there is mucosa with multiple minor salivary glands called labial glands.[@b8-ccid-7-191],[@b9-ccid-7-191]
The red color of the lip margin is due to a translucence from the blood vessels present in the underlying connective tissue papillae. These superficial papillae are rich in blood capillaries and sensory nerve endings, which render the lip highly vascular as well as highly sensitive. No sweat glands or hair follicles are present in the free red margin of the lip.
The muscular basis of the lip and the perioral region consists of the orbicularis oris muscle (OOM), which is subdivided into two parts: the superficial part and the deep part, the deep part itself consists of pars peripheralis and pars marginalis.
The functions of the lips in humans are various: the archaic parts account for sphincter function, food intake (catching food, breaking food, eating, moistening, and tasting), and breathing, whereas the phylogenetically newer parts serve communication through facial expression and speech.
The submuscular aponeurotic system is the anchoring of the OOM and perioral mimic muscle to it cranially, as well as to the platysma caudally. This has been studied by Thaller et al in ten fresh cadaver heads and twelve rhesus monkey fetuses. The authors described this distinct fibromuscular layer and its relation to fascia and muscles, and reported that it serves as a covering as well as connector and anchor between muscles and skin, maintaining tension in the face.[@b10-ccid-7-191]
The perioral area is not only well circulated by lymphatic and blood vessels, but also by numerous mimic muscle fibers accounting for vivid facial expression. These mimic muscle fibers intertwine and extend from the superficial part of the OOM into the platysma and into the submuscular aponeurotic system, the superior levator labii muscle, and the zygomaticus major muscle. They create the suspension of the upper and lower lip in the lower face.
D'Andrea and Barbaix, in a 2006 study of the perioral muscles, conducted anatomic research in 40 embalmed Caucasian specimens and compared them to magnetic resonance images of ten live subjects. They demonstrated the relations of the orbicularis oris, the perioral and buccinator muscles and the muscles' path.[@b11-ccid-7-191] A further three-dimensional magnetic resonance imaging study of the lip muscles by Olszewski et al in 2009 revealed the different structures such as adipose tissue, muscle fibers, vessels, nerves, and glandular structures. The authors described a novel, noninvasive, in vivo method for segmenting and digitally reconstructing facial muscles and demonstrated clearly the direction of muscle fibers and structures.[@b12-ccid-7-191] The anatomic structures essential for functions such as lip closure and articulation have been well described,[@b12-ccid-7-191] along with differences in function between normal and cleft palate lips. By observing important structures in the development of the OOM, Mooney et al,[@b13-ccid-7-191] could perform a three-dimensional reconstruction of this muscle.[@b13-ccid-7-191]
McAlister et al demonstrated in 30 Caucasian and 24 Asian dental students the differences in the thickness of the lip levator musculature and showed an improved technique of ultrasonographic imaging of the lip levator musculature.[@b14-ccid-7-191] They investigated the zygomaticus major muscle and the levator labii superioris muscle in order to describe orthodontic results. They characterized smile lines as "high", "medium", or "low" in terms of the relation of lip to teeth and measured the thickness of the superior levator labii and major zygomaticus muscles. They found that females had a higher smile line and a thicker major zygomaticus muscle than males. The OOM was not investigated in this study.[@b14-ccid-7-191]
The philtrum, a very specific and characteristic area with many intertwining structures, was investigated by Briedis and Jackson,[@b15-ccid-7-191] who performed cadaver dissections and histological studies on the philtral region of eight adults and four fetal lips to elucidate the detailed anatomy of the OOM in the intact normal lips. They described that the OOM in pars peripheralis and pars marginalis are confined to the lip and vermilion. The muscles were found to continue to the buccinator muscle profoundly and to the facial muscles superficially, which has to be taken into account in cleft lip repair.[@b15-ccid-7-191]
Overall, these studies show interindividual variation in the perioral anatomy, which must to be taken into account and studied well when intervening in this area. Furthermore, the blood supply of the lips and perioral region is very strong and can thus result in large hematomas when injecting this area, as demonstrated in the anatomic study by Pinar et al[@b5-ccid-7-191] and in the book *Surgical Anatomy of the Face*.[@b6-ccid-7-191] These publications are well worth studying before injecting the modiolar region.
Injectable fillers: injection techniques, histology, and complications
----------------------------------------------------------------------
A variety of filler materials with different molecular chain length and linkage with characteristics of hydration and degradation are available, and can be used to achieve different and natural-looking results.
In a recent in vitro/ex vivo publication, Eversole et al described the histopathologic features of soft tissue filler reactions of the lips. They state that those reactions are rare and that exact numbers of incidences are unknown.[@b16-ccid-7-191] They investigated instances of perioral and labial foreign body reactions to a variety of injectable soft tissue fillers with the objective to identify histopathologic characteristics, allowing the pathologist to identify the injected materials. Eversole et al could identify the different materials causing foreign body reactions in vitro and concluded that soft tissue-filler reactions, or similarly foreign body-host-reactions in conjunction with the morphology of the foreign materials themselves, can be differentiated from one another microscopically.[@b16-ccid-7-191] One can further conclude that even well-tolerated injected materials have characteristic histopathologic morphologies.
An extensive review on adverse events by Requena et al in 2011 described several different filler materials and their possible indications and adverse reactions in clinical and histopathological images of tissue biopsies after surgical removal of the filler. They state that a wide variety of cosmetic fillers are available worldwide, but the ideal filler is still missing, because all fillers known today may cause adverse reactions. Requena et al show dramatic photographs and microscopic images after augmentation, and stress the importance of thorough knowledge of anatomy before injecting.[@b4-ccid-7-191]
Investigating a new injectable hyaluronic acid (HA) filler, Juvéderm^®^ Volbella™ (Allergan, Inc., Irvine, CA, USA) for use in the perioral area in 2012, Eccleston and Murphy, in a 12-month prospective, multicenter, open-label study in 60 subjects, showed the results of injecting the perioral area. They investigated the "fullness goal achievement", the look and feel of the lips, and the patients' satisfaction with the effects of treatment. Juvéderm^®^ Volbella™ injectable HA was demonstrated to be well tolerated and to provide a smooth and natural improvement in lip fullness that lasted for up to one year.[@b17-ccid-7-191] This is the only prospective, long-term study in HA augmentation of the lips. This filler was not yet available commercially at the time this present study was conducted.
Sarnoff et al[@b18-ccid-7-191] and Sarnoff and Gotkin[@b19-ccid-7-191] describe in two publications a step-by-step guide to lip augmentation as well as which filling agents to use by comparison of filling agents available at the time for lip augmentation (Restylane^®^ \[Galderma S.A., Lausanne, Switzerland\], Belotero^®^ \[Merz Pharmaceuticals GmbH, Frankfurt, Germany\], and Juvéderm^®^ \[Allergan Inc.\]). Indications for injections were to augment the lips, to correct perioral rhytides, and to enhance overall lip appearance. The authors underline that the goal for upper lip augmentation is to create a form that harmonizes with the patient's unique features, taking into account age and ethnicity; the goal for the lower lip is to create bulk, greater prominence, and projection of the vermilion.[@b18-ccid-7-191],[@b19-ccid-7-191]
In 2007, Ali et al published a review on perioral rejuvenation and lip augmentation.[@b20-ccid-7-191] They state that aging of the face is a process of atrophy, most noticeable in the perioral region. They discuss rejuvenation of the perioral region, including with fillers, surgery, and facial resurfacing, as correction for this aging process. Detailed techniques for each of the approaches are outlined. Composition of the various fillers is discussed by Ali et al in conjunction with their respective outcomes and duration of effect as documented by photography.[@b20-ccid-7-191] No special attention was paid to complications. This, in contrast, is the main topic of Weinberg and Solish's article on complications in hyaluronic fillers. They describe the most frequent and serious complications (eg, hematoma, infection, granulomata), their prevention, and treatment.[@b3-ccid-7-191]
In a meta-analysis in 2013, Cohen et al evaluated 53 publications about HA used for aesthetic soft tissue augmentation.[@b21-ccid-7-191] They found out of 53 primary clinical reports, that the highest-quality efficacy evidence was for injecting the nasolabial folds in 10 randomized, blinded, split-face, comparative trials. Several randomized, blind trials supported treatment of the glabella, lips, and hands. Lower-level evidence (from studies with nonrandomized, open-label, or retrospective designs) was recorded for the nasojugal folds (tear troughs), upper eyelids, nose, infraorbital hollows, oral commissures, marionette lines, perioral rhytides, temples, and cheeks. Common adverse events across anatomic areas were pain, bruising, swelling, and redness. Serious adverse events were uncommon (eight events in eight patients of 4,605 patients in total) and were considered to be unrelated (seven events) or probably unrelated (one event) to treatment.[@b21-ccid-7-191] Cohen et al concluded that the efficacy and safety of small- and large-gel-particle HA are well established for nasolabial folds; evidence for the glabella, lips, and hands is more limited. Preliminary reports in other anatomic regions suggest efficacy without major complications.[@b21-ccid-7-191]
Sommer states that the so-far described injection techniques do not take the anatomic structures into account accordingly, and postulates injecting the vermilion from peripherally transcutaneously in order to avoid major bruising.[@b22-ccid-7-191]
Carruthers and Carruthers, in 2005, published an article on facial sculpting and tissue augmentation with filler material.[@b23-ccid-7-191] They also developed a grading system for lip fullness to objectify the augmentation goal and to provide aesthetic guidelines and described their favorite techniques[@b24-ccid-7-191] ([Table 1](#t1-ccid-7-191){ref-type="table"}). They state that there are five grades of lip fullness, which range from very thin (0) to full (4) on a 5-point photonumeric rating scale designed to objectively quantify the three-dimensional fullness of the lip.[@b23-ccid-7-191],[@b24-ccid-7-191] This scale does not take the relation between upper and lower lip into account.
Sclafani, in 2005, described techniques of soft tissue augmentation for the management of the aging perioral complex. There are various techniques for augmenting the perioral region, such as tunnel, stamp, and depot techniques. Sclafani stresses that careful analysis of this area and appropriate treatment can harmonize these areas and produce a globally aesthetic result without surgery.[@b26-ccid-7-191]
Aim of the current study
------------------------
It was our aim to define the exact position of HA in the lips and the perioral region immediately after injection and 18 days post-injection.
We also wanted to record possible complications. Although avoidance of complications is preferred, it is incumbent on the physician to have a detailed understanding of the perioral anatomy and pathophysiology as well as how to prevent and manage complications.
Materials and methods
=====================
There were ten patients initially included in this prospective pilot study. All patients presented for aesthetic augmentation of the lips and perioral area, some also demanded hydration of the lips to reduce minimal wrinkles. One patient dropped out of the study for not showing up for the follow-up exam, so nine patients were included and evaluated. After thorough history taking, photo documentation was performed. A pre-injection ultrasonography, as well as informed written consent, were obtained.
Included in this pilot study were volunteers who agreed to have the lower third of the face augmented after informed, written consent was obtained. Subjects who had undergone augmentation in the past 6 months, who had known allergies/intolerance against lidocaine, and who suffered from coagulopathies were excluded. Dysmorphophobic and psychologically unstable persons were also excluded from undergoing injection augmentation.
The patients were injected with Juvéderm^®^ Ultra III^®^ and Ultra Smile^®^ (Allergan, Inc.), and seven were additionally injected with unlinked HA (Juvéderm^®^ Hydrate^®^; Allergan, Inc.). The exact amounts, locations, and injection techniques were recorded on a standardized documentation sheet and table (see [Tables 2](#t2-ccid-7-191){ref-type="table"} and [3](#t3-ccid-7-191){ref-type="table"}).
Imaging
-------
First, photographic documentation of the face -- static and in motion -- was performed. The patients were photographed in the following positions/with the following instructions: static/at rest looking relaxed and straight forward; closing the eyes as if sleeping; closing the eyes firmly as if the sun were shining; kissing; smiling; showing teeth; raising eyebrows to wrinkle the forehead; raising the nasal tip as when sniffing; and looking angry.
Then, high-resolution ultrasonography of the lower third of the face was conducted. A Hitachi HI VISION Avius^®^ ultrasound system with linear scanner (frequency range: 6--14 MHz; Hitachi Ltd, Tokyo, Japan) supplied images of the OOM and the surrounding lip tissue. Blood vessels were detected by color Doppler flow mapping with the same device in Doppler flow mode. The amount of pressure applied on the sonography head could not be standardized and thus structures were recorded in a dynamic way with varying pressures and flow documentations. A gel pad was used (Aquaflex^®^ 1 cm; Parker Laboratories, Fairfield, NJ, USA) for optimal ultrasonographic transmission.[@b27-ccid-7-191]
Last, slit-light optical coherence tomography (SL-OCT) of the lip surface was performed to objectify even minimal wrinkles of the lips. The SL-OCT used was a Spectralis^®^ (Heidelberg Engineering, Heidelberg, Germany) with the Slit lamp BD 900^®^ (Heidelberg Engineering), and background illumination with Eco-lite EL 01/02 (Haag-Streit AG (Koeniz, Switzerland). The wave length of the diode laser beam was approx. 1310 nm. The resulting A scans had a pattern size of 15×7 mm. We scanned both the upper and lower lip, each in 3 positions (medial, right lateral, left lateral). Thus, we obtained 6 A scans of each patient.
The examinations were performed before, immediately after, and 18 days after injection in five patients. Due to time constraints, four patients were examined only once on day 1.
The dates of investigations were chosen as the day of the injection (before and immediately after the injection), as well as 18 days after the injection for convenience of patients and examiners.
### Injection techniques
There are various techniques for augmenting tissue, such as depot technique (to augment a deep tissue on a spot), tunnel injection (by outlining a structure in linear, horizontal underlay technique), and stamp injection (injecting a deep depot, lifting the needle through the tissue to the surface while injecting, creating a vertical depot). The aim of each of these is to achieve maximum results with the least material and injury, for augmentation and wrinkle reduction with a natural-looking result.
Only sharp and non-blunt cannulas were used for injection, paramedian at both upper and lower lip in depot technique, and around the vermilion border to enhance it in tunnel technique. Further, marionette lines and deep nasolabial lines were evened out by tunnel and stamp injection techniques. All of the above were performed with Juvéderm^®^ Ultra III^®^ and Ultra Smile^®^.
Juvéderm^®^ Hydrate^®^ was deposited subcutaneously in the philtrum/upper lip area to even out small rhytides and augment the subcutaneous tissue.
Immediately after injection, patients were provided with cold (10°C), moist pads to cool their injected perioral region.
A combination of these fillers was chosen to achieve best results according to the indications. Using a higher cross-linked, longer HA (Ultra III^®^ and Ultra Smile^®^ \[ Allergen Inc.\]) resulted in longer lasting, more stable results and hygroscopic effects, whereas the unlinked HA (Juvéderm^®^ Hydrate^®^) had the most immediate rhytide-reducing effects.
All fillers were used in combination and recorded in the patient's chart (see [Table 3](#t3-ccid-7-191){ref-type="table"}).
Results
=======
Out of all 21 applicant patients, nine females, aged 46 to 64 years (mean 55.0 years, range 41.2--66. 1 years), were eligible to participate in this pilot study. Three women were injectable-naïve and six had previously received injectable filler augmentations of the lips and face. All were otherwise healthy, taking no medication on a regular basis (except for hormone substitution) and with no known allergies.
[Figure 1](#f1-ccid-7-191){ref-type="fig"} demonstrates a systematic B-mode ultrasonography of the lips, which was performed in every patient and included:
- static sonogram at five defined points;
- dynamic sonogram (from midline to right and from midline to left modiolus);
- and Doppler sonography for identification of flow and vessels.
The injected material distributed well within the lip tissue, and no embolism or thrombosis occurred in our investigated cases. The location and position of the fillers was confirmed by ultrasound in the anatomically intended spots. Some of the injected material, intended for volume augmentation of the lip, was deposited in the deep part (pars peripheralis) of the OOM. The injected material came up to 1 mm from the arteries and veins of the lips.
The sonography ([Figure 3](#f3-ccid-7-191){ref-type="fig"}) shows how the HA depots lie in the OOM: one can differentiate lacunar and laminar depots, depending on the injection technique (round vesicles on depot technique, laminar and lacunar in stamp and tunnel technique). It can clearly be differentiated in the tissue directly post-injection, and merges with the tissue over time. [Figure 3](#f3-ccid-7-191){ref-type="fig"} shows sonographic images before and immediately after injection in a paramedian, sagittal B-mode scan of the upper lip. The muscle appears broader and thicker by the injected depots (64-year-old female). Depicted is the OOM pars superficialis and pars profunda, the latter of which is divided into pars peripheralis and pars marginalis (the shape of the muscle is akin to a hockey stick). The merging and distribution of HA with the injected tissue are ultrasonographically shown by a decontouring of depots.
The hygroscopic effect was indirectly visible by the increased volume and the decreased rhytides of the lips, as shown by OCT. The fillers were injected and thus deposited in different strata/layers (intramuscular, subcutaneously, in between the small salivary glands), which was shown by ultrasonography (see [Figure 3B](#f3-ccid-7-191){ref-type="fig"}).
In our nine patients, no compression of lip structures such as vessels or nerves occurred, nor did any severe complications such as major bleeding, infection, or thromboembolism result from injection. The side effects disappeared within one week ([Figure 2](#f2-ccid-7-191){ref-type="fig"}).
Clinically and in self-rating of the patients, the lips were smoother, more moist, and less wrinkled, which could be demonstrated and objectified by OCT as seen in [Figure 4](#f4-ccid-7-191){ref-type="fig"}. The OCT imaging could support the scientific insight that HA saves water and can thus contribute to wetter/more hydrated and fuller lips as demonstrated by significantly reduced rhytides and scaling. The amount of augmentation achieved to a natural-looking result was rated by the Carruthers' scale and is shown in [Table 4](#t4-ccid-7-191){ref-type="table"}.
Discussion
==========
There are no limits or guidelines yet as to how deep and how much substance may be injected in the human perioral region. Some authors state that the injection of hyaluronic fillers should be in the superficial part of the OOM.[@b28-ccid-7-191] Further studies must quantify the exact depth and the amount of injected filler materials by simultaneous ultrasonography during injection.
The method of ultrasonography applied in this study has an intrinsic problem, since, depending on the pressure applied with the sonic head, the structures appear differently. As such, the combination of static and dynamic investigation was essential and resulted in a good overview of the position of the injected material.
If small salivary glands are injected, retention cysts and sialadenitis can occur, but, when using resolvable HA, this has not yet been described.
It is crucial to know the exact anatomy to minimize complications and risks when injecting fillers. The position of injectables depends on the technique used, such as depot, stamp/tower, or tunnel.[@b29-ccid-7-191] Adverse events such as bruising or thrombosis can be avoided by ultrasonographic visualization of vessels before injection and during injection control and thus early counter-action of thrombosis or compression of structures. One should always have hyaluronidase as a rescue medication at hand. When in doubt, the exact location of injected fillers should be controlled by Doppler ultrasound to avoid injury of arteries or veins; however, this is nearly impossible in a two-handed injection process.
More superficial injections could avoid injuring vessels; however, this is not always aesthetically advisable[@b22-ccid-7-191] and does not achieve a good amount/result ratio when trying to achieve the best effect with the least material.
Further studies to follow up at later time points after injection (eg, 3 and 6 months) would be beneficial and are in planning.
In our opinion, intramuscular injections are not absolute contraindications as stated by Lemperle et al[@b28-ccid-7-191],[@b30-ccid-7-191] and Lemperle and Duffy.[@b31-ccid-7-191] Intramuscular injections may sometimes be necessary for an effective augmentation and can occur unnoticed when no ultrasonographic control is performed.
Conclusion
==========
Common complications of the perioral region after HA augmentation such as hematoma and pain can occur in the first post-injection days, but severe bleeding, nerve injuries, thromboembolism, or even blanching, are rare.
Hyaluronidase should always be available in order to immediately be able to dissolve the filler if any crucial structures such as arteries or larger veins are compressed, injected, or blocked in order to avoid major complications with persisting damage.
Generally, HA injection is a safe and minimally invasive method by which to augment the perioral region, if one is aware of the numerous crucial anatomic structures.
It would be advantageous to locate crucial structures and the position of the injected material post-injection by ultrasound in order to avoid major complications, in case of any doubt of the position of the injected material arises, so you can "know where your fillers go" by ultrasonographic control. However, such high-end ultrasonography is not always possible, and thus such a thorough investigation may not be feasible for routine clinical application. It is desirable to raise awareness of these technical possibilities, such as high resolution ultra-sonography and OCT, and to increase their use wherever available.
Parts of the data acquired in this study were presented at the Meeting of the European Academy of Facial Plastic Surgery (EAFPS) in Rome, May 2012, and won the first prize for best e-poster presentation at the German Academy meeting for Aesthetic Surgery (GÄCD in Freiburg, 2012, and Munich, 2013) and at the Austrian Academy meeting for Otorhinolaryngology 2012 in St Pölten.
**Disclosure**
Injectable fillers were provided by Allergan Inc., and Dr Vent received travel support from Allergan Inc. for presenting these data at GÄCD in 2012, Freiburg, Germany. The authors report no other conflicts of interest in this work.
{#f1-ccid-7-191}
{#f2-ccid-7-191}
{#f3-ccid-7-191}
{#f4-ccid-7-191}
######
Carruthers' et al[@b24-ccid-7-191] lip fullness grading scale
Grade 0 1 2 3 4
----------------------------------------- --------------- ---------------------- -------------------- ----------------------- -----------------
Description Very thin Thin Moderately thick Thick Very thick
Volume/vermilion Small Large
Relation to SL (from columella to chin) Far behind SL Moderately behind SL Slightly behind SL Lower lip touching SL In line with SL
**Abbreviation:** SL, Steiner line.
######
Indications, locations, and techniques for fillers
Juvéderm Ultra III^®^ (0.8 mL syringe) and Ultra Smile^®^ (0.55 mL syringe) Juvéderm Hydrate^®^ (1 mL syringe)
----------------------------------------------------------------------------- ----------------------------------------------------------------------------------------------------
Augmentation of lip/vermilion (stamp/depot technique) Rehydration of dry lips[\*](#tfn3-ccid-7-191){ref-type="table-fn"}
Contouring of vermilion border (funnel technique) Small rhytides[\*](#tfn3-ccid-7-191){ref-type="table-fn"}
Lifting of modiolus (funnel technique) Augmentation of philtrum[\*](#tfn3-ccid-7-191){ref-type="table-fn"}
Flattening of nasolabial fold (funnel and stamp techniques) Augmentation of upper lip from nose to vermilion border[\*](#tfn3-ccid-7-191){ref-type="table-fn"}
**Note:** Juvéderm^®^ products: Allergan, Inc., Irvine, CA, USA.
Intracutaneous injection, superficial depots.
######
Amounts of filler material used and indications per patient
Patient Juvéderm^®^ Ultra III^®^ Juvéderm Ultra Smile^®^ Juvéderm Hydrate^®^
------------------ ----------------------------------------------------------------------- --------------------------------------------------------------------- -----------------------------
1\. 48 years old 0.8 mL, marionette lines 0.55 mL UL volume, vermilion border 0
2\. 45 years old 2×0.8=1.6 mL, vermilion border, LL volume 0 1 mL, UL rhytides
3\. 41 years old 2×0.8=1.6 mL, nasolabial fold, marionette lines 0.55 mL, vermilion border 1 mL, UL rhytides, philtrum
4\. 62 years old 2×0.8=1.6 mL, nasolabial fold, marionette lines 0.55 mL, vermilion border 1 mL, UL rhytides
5\. 59 years old 0.8 mL, nasolabial fold, marionette lines 0 1 mL, UL rhytides
6\. 66 years old 0.8 mL, nasolabial fold, modiolus 0.55 mL, vermilion border, UL volume 1 mL, UL rhytides
7\. 46 years old 0 3×0.55=1.65 mL, vermilion border, UL/LL volume, and nasolabial fold 1 mL, UL rhytides
8\. 64 years old 3×0.8=2.4 mL, nasolabial fold, vermilion border, and UL/LL lip volume 0 0
9\. 63 years old 2×0.8=1.6 mL, nasolabial fold, chin/marionette lines 2×0.55=1.1 mL, vermilion border, volume depot UL/LL 0
**Note:** Juvéderm^®^ products: Allergan, Inc., Irvine, CA, USA.
**Abbreviations:** LL, lower lip; UL, upper lip.
######
Carruthers' scale grading of patients' lip fullness before and after injection
Patient Before injection After injection
------------------ ------------------ -----------------
1\. 48 years old 0 1
2\. 45 years old 1 3
3\. 41 years old 0 2
4\. 62 years old 0 2
5\. 59 years old 3 3
6\. 66 years old 0 1
7\. 46 years old 1 3
8\. 64 years old 0 1
9\. 63 years old 0 1
**Notes:** The lip fullness can be classified into five fullness grades, ranging from very thin (0) to full (4) on a 5-point photonumeric rating scale. See [Table 3](#t3-ccid-7-191){ref-type="table"} for amounts of hyaluronic acid injected.
|
{
"pile_set_name": "PubMed Central"
}
|
<?php
namespace Oro\Bundle\VisibilityBundle\Async\Visibility;
use Doctrine\Common\Persistence\ManagerRegistry;
use Doctrine\DBAL\Exception\RetryableException;
use Doctrine\ORM\EntityManagerInterface;
use Doctrine\ORM\EntityNotFoundException;
use Oro\Bundle\ProductBundle\Entity\Product;
use Oro\Bundle\VisibilityBundle\Async\Topics;
use Oro\Bundle\VisibilityBundle\Entity\VisibilityResolved\ProductVisibilityResolved;
use Oro\Bundle\VisibilityBundle\Visibility\Cache\CacheBuilderInterface;
use Oro\Bundle\VisibilityBundle\Visibility\Cache\ProductCaseCacheBuilderInterface;
use Oro\Component\MessageQueue\Client\TopicSubscriberInterface;
use Oro\Component\MessageQueue\Consumption\MessageProcessorInterface;
use Oro\Component\MessageQueue\Transport\MessageInterface;
use Oro\Component\MessageQueue\Transport\SessionInterface;
use Oro\Component\MessageQueue\Util\JSON;
use Psr\Log\LoggerInterface;
/**
* Resolves visibility for a product when its category is changed.
*/
class ProductProcessor implements MessageProcessorInterface, TopicSubscriberInterface
{
/** @var ManagerRegistry */
private $doctrine;
/** @var CacheBuilderInterface */
private $cacheBuilder;
/** @var LoggerInterface */
private $logger;
/**
* @param ManagerRegistry $doctrine
* @param LoggerInterface $logger
* @param CacheBuilderInterface $cacheBuilder
*/
public function __construct(
ManagerRegistry $doctrine,
LoggerInterface $logger,
CacheBuilderInterface $cacheBuilder
) {
$this->doctrine = $doctrine;
$this->logger = $logger;
$this->cacheBuilder = $cacheBuilder;
}
/**
* {@inheritDoc}
*/
public static function getSubscribedTopics()
{
return [Topics::CHANGE_PRODUCT_CATEGORY];
}
/**
* {@inheritdoc}
*/
public function process(MessageInterface $message, SessionInterface $session)
{
$body = JSON::decode($message->getBody());
if (!isset($body['id'])) {
$this->logger->critical('Got invalid message.');
return self::REJECT;
}
/** @var EntityManagerInterface $em */
$em = $this->doctrine->getManagerForClass(ProductVisibilityResolved::class);
$em->beginTransaction();
try {
if ($this->cacheBuilder instanceof ProductCaseCacheBuilderInterface) {
$productIds = array_unique((array) $body['id']);
$products = $this->getProducts($productIds);
foreach ($products as $product) {
$this->cacheBuilder->productCategoryChanged($product, $body['scheduleReindex'] ?? false);
}
}
$em->commit();
} catch (\Exception $e) {
$em->rollback();
$this->logger->error(
'Unexpected exception occurred during Product Visibility resolve by Product.',
['exception' => $e]
);
if ($e instanceof RetryableException) {
return self::REQUEUE;
}
return self::REJECT;
}
return self::ACK;
}
/**
* @param int[] $productIds
*
* @return Product[]
*
* @throws EntityNotFoundException If all products have not been found
*/
private function getProducts(array $productIds): array
{
/** @var Product[] $products */
$products = $this->doctrine->getManagerForClass(Product::class)
->getRepository(Product::class)
->findBy(['id' => $productIds]);
if (count($productIds) !== count($products)) {
$foundIds = array_map(
static function (Product $product) {
return $product->getId();
},
$products
);
$notFoundProductsIds = array_diff($productIds, $foundIds);
$this->logger->warning(
'The following products have not been not found when trying to resolve visibility',
$notFoundProductsIds
);
if (!$products) {
throw new EntityNotFoundException(
sprintf(
'Products have not been found when trying to resolve visibility: %s',
implode(', ', $notFoundProductsIds)
)
);
}
}
return $products;
}
}
|
{
"pile_set_name": "Github"
}
|
1. Field of the Invention
The present invention relates to a capacitance-based sensor, which is embedded in a vehicle seat, and also relates to an occupant sensing system, which determines an occupant state of the seat based on an output of the capacitance-based sensor.
2. Description of Related Art
One type of occupant sensing system includes a capacitance-based sensor and an occupant sensing electronic control unit (ECU). The capacitance-based sensor senses a disturbance in a weak electric field generated by an electrode and outputs the sensed result as the corresponding electric current or electric voltage (see, for example, Japanese Unexamined Patent Publication No. H11-271463).
Specifically, in a state of an empty seat where no occupant is present in the seat, the air is interposed between two electrodes of the capacitance-based sensor. In another state where a child restraint system (CRS) is installed in the seat, the CRS is interposed between the two electrodes of the capacitance-based sensor. In yet another state where an occupant is present in the seat, a body of the occupant is interposed between the two electrodes of the capacitance-based sensor.
Here, a dielectric constant of the air is about 1. A dielectric constant of the CRS is about 2 to 5 although it may vary depending on a material of the CRS. Furthermore, a dielectric constant of the human body is about 50. As illustrated above, the dielectric constant of the air, the dielectric constant of the CRS and the dielectric constant of the human body differ from one another. Thus, a capacitance between the two electrodes of the capacitance-based sensor varies depending on the type of the interposed object, which is interposed between the two electrodes.
Based on the differences in the capacitance, the occupant sensing ECU performs the occupant determination. Specifically, the occupant sensing ECU determines whether the seat is empty, whether the CRS is installed in the seat and/or whether an adult is present in the seat. An air bag ECU determines enablement/disablement of deployment of an air bag based on the result of the determination of the occupant sensing ECU. Specifically, in the state of the empty seat or in the state of the CRS installed in the seat, the deployment of the air bag is disabled. On the other hand, in the state where the adult is present in the seat, the deployment of the air bag is enabled.
The capacitance, which is measured with the capacitance-based sensor in the state where the occupant is present in the seat, includes the capacitance, which is generated upon seating of the occupant in the seat, and the capacitance, which has been present since the empty state of the seat. This point will be described with reference to FIG. 34. FIG. 34 is a diagram schematically showing a circuit structure of a previously proposed capacitance-based sensor. In FIG. 34, “Cb” is a capacitance between an electrode 120 and a vehicle ground, i.e., GND (or a seat frame that is grounded to the vehicle ground) through a human body, and “Co” is a capacitance (an empty seat capacitance), which has been present between the electrode 120 and the vehicle GND since the empty state of the seat. An electric current sensing device (an electric current sensing circuit) 131 measures the capacitance as a sum of the capacitance Cb and the capacitance Co. The capacitance (the empty seat capacitance) Co between the occupant sensing electrode 120 and the vehicle GND constitutes a relatively large ratio in the entire capacitance (the sum of the capacitance Cb and the capacitance Co), and variations in the empty seat capacitance Co directly causes variations in the entire capacitance to deteriorate the level of the accuracy in the occupant determination. Furthermore, the adjustment of the variation in the measured capacitance of the capacitance-based sensor in each vehicle causes a large increase in the factory work load.
Furthermore, with reference to FIG. 35, in the case of the occupant sensing system recited in Japanese Unexamined Patent Publication No. H11-271463, an occupant sensor 514 is provided on a top surface of a seat bottom 510 of a passenger seat. The occupant sensor 514 includes a first electrode 512 and a second electrode 513. An electric field is created between the first electrode 512 and the second electrode 513. When an occupant is seated on the seat bottom 510, a capacitance between the first electrode 512 and the second electrode 513 changes to cause a change in the electric current between the first electrode 512 and the second electrode 513. By measuring the change in this electric current, presence of the occupant on the seat bottom 510 is sensed.
However, there is provided only the single set of electrodes 512, 513. Thus, it is not possible to cover various seating patterns of the occupant. For example, when the occupant is displaced from the single set of electrodes 512, 513, the presence of the occupant on the seat bottom 510 cannot be effectively sensed.
|
{
"pile_set_name": "USPTO Backgrounds"
}
|
/*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation; either version 2 of the License, or
* (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU Library General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, write to the Free Software
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
*
* SunliteFactory.h
* The WidgetFactory for SunLite widgets.
* Copyright (C) 2014 Simon Newton
*/
#ifndef PLUGINS_USBDMX_SUNLITEFACTORY_H_
#define PLUGINS_USBDMX_SUNLITEFACTORY_H_
#include "libs/usb/LibUsbAdaptor.h"
#include "ola/base/Macro.h"
#include "plugins/usbdmx/Sunlite.h"
#include "plugins/usbdmx/WidgetFactory.h"
namespace ola {
namespace plugin {
namespace usbdmx {
/**
* @brief Creates SunLite widgets.
*/
class SunliteFactory : public BaseWidgetFactory<Sunlite> {
public:
explicit SunliteFactory(ola::usb::LibUsbAdaptor *adaptor)
: BaseWidgetFactory<Sunlite>("SunliteFactory"),
m_adaptor(adaptor) {}
bool DeviceAdded(
WidgetObserver *observer,
libusb_device *usb_device,
const struct libusb_device_descriptor &descriptor);
private:
ola::usb::LibUsbAdaptor* const m_adaptor;
// The product ID for widgets that are missing their firmware.
static const uint16_t EMPTY_PRODUCT_ID;
// The product ID for widgets with the firmware.
static const uint16_t FULL_PRODUCT_ID;
static const uint16_t VENDOR_ID;
DISALLOW_COPY_AND_ASSIGN(SunliteFactory);
};
} // namespace usbdmx
} // namespace plugin
} // namespace ola
#endif // PLUGINS_USBDMX_SUNLITEFACTORY_H_
|
{
"pile_set_name": "Github"
}
|
SIT Orientation Videos
Event Details:
Date: 23 - 25 August 2019 (Open to all programmes).
Time: 8am first day to 11am last day.
Registration - Open from 15 May to 30 June 2019
Please note registration is based on a first come first served basis due to space constraint.
Registration link will be sent to your email. Do look out for it within a week upon accepting your programme offer.
|
{
"pile_set_name": "Pile-CC"
}
|
Ptah
Ptah is a fast, fun, open source high-level Python web development environment. Ptah is built on top of the Pyramid web framework. Ptah’s goal is to make developing interactive web sites and applications fun. Ptah aims to fill a void in the Pyramid eco-system, a “full stack” environment which is well integrated and provides opinions (forms, management ui, models, etc).
Ptah is loosely affiliated with the Pyramid, Django, Drupal and Zope/Plone communities.
Most documentation requires Ptah 0.3 or greater.
You can read the ptah documentation on-line at http://ptahproject.readthedocs.org.
Requirements Python 2.6+ or Python 3.2+
virtualenv
Note for Windows Users On Windows virtualenv/bin will be virtualenv/Scripts besides this difference everything else below is the same.
Grab the release If you do not want to faff about with source, cloning repos, etc. Just grab the latest released version of ptah. $ /path/to/virtualenv/bin/pip install ptah
Ptah from source If you want the latest and greatest you need to grab code from source. clone ptah from github and then install it: $ /path/to/virtualenv/bin/python setup.py develop then run the tests: $ /path/to/virtualenv/bin/python setup.py test
An empty project Let’s generate a empty project using the ptah_starter scaffolding. You can start from there: /path/to/virtualenv $ bin/pcreate -t ptah_starter myapp /path/to/virtualenv $ cd myapp /path/to/virtualenv/myapp $ ../bin/python setup.py develop /path/to/virtaulenv/myapp $ ../bin/pserve settings.ini --reload Open your browser to http://localhost:6543/ if you want examples that do more such as demonstrating editing models and authentication. Check out the examples.
Examples There are several example applications ready for you to install and see Ptah in action. You can find them in the examples repository at github. https://github.com/ptahproject/examples
Support and Documentation Ptahproject google groups/mailing list, Ptahproject Google Groups On irc, use the freenode network and find us on channels, #ptahproject and #pyramid. Documentation can be found in docs directory. You can also see it online at http://ptahproject.readthedocs.org/ Report bugs at Ptahproject @ Github
Known Issues On some versions of Ubuntu you may get Python 2.7 exiting stating it has “Aborted.” There is a bug in ctypes on that particular Ubuntu platform.
|
{
"pile_set_name": "OpenWebText2"
}
|
Adrian College Congratulates Graduates of Winter Commencement
Adrian, MICH – On behalf of President Jeffrey R. Docking, Adrian College is pleased to present its most recent graduates. In an intimate ceremony, with family, friends and members of the College community present, 48 students received their diplomas. Graduates are listed below in alphabetical order, along with their hometowns and degrees earned.
Traditionally, “summa cum laude,” “magna cum laude,” and “cum laude” are used on diplomas to grant three special honors for grades above the average. Summa cum laude means “with highest praise” and represents a graduate with a cumulative grade point average of 3.8 and above. Magna cum laude means “with great praise” and represents a graduate with a cumulative grade point average between 3.65-3.79. Cum laude means “with honor” and represents a graduate with a cumulative grade point average between 3.5-3.64.
|
{
"pile_set_name": "Pile-CC"
}
|
Background
==========
Attention-deficit/hyperactivity disorder (ADHD) is one of the most common neurobehavioral psychiatric disorders that afflicts children \[[@B1]\], with a reported prevalence of 2.4% to 19.8% worldwide \[[@B2]\] using the criteria from the *Diagnostic and Statistical Manual of Mental Disorders*, *Fourth Edition*(DSM-IV) from the American Psychiatric Association \[[@B3]\]. Two Canadian studies of children and adolescents, using earlier diagnostic criteria to examine ADHD prevalence, estimated a prevalence of 6.3% in an Ontario study of participants (aged 4 to 16 years) \[[@B4]\], and 3.3% to 8.9% in a comparable population (aged 6 to 14 years) in Quebec \[[@B5]\].
Stimulants have long been used to treat ADHD symptoms. The Texas Consensus Conference Panel on Pharmacotherapy of Childhood ADHD algorithm \[[@B6]\] considered psychostimulants as first-line pharmacotherapy treatments for ADHD; However, the Canadian ADHD Resource Alliance (CADDRA) guidelines consider long-acting stimulants and atomoxetine as first-line agents in the management of ADHD \[[@B7]\]. The stimulant types most commonly used in ADHD treatment are methylphenidate (MPH) and amphetamine. These have similar subjective effects \[[@B8]\] yet differ in their mechanisms of action\--MPH is a dopamine and norepinephrine reuptake inhibitor, while amphetamines have additional presynaptic activity\--stimulating the release of dopamine, norepinephrine, and serotonin \[[@B9]\]. Although both are considered efficacious, a meta-analysis of 23 studies comparing the efficacy of immediate-release (IR) formulations of MPH and amphetamine in the treatment of children with ADHD revealed small but statistically significant differences in favor of amphetamine \[[@B10]\]. A comparative review of controlled crossover studies \[[@B11]\] found that clinical response rates for IR formulations of MPH and amphetamine ranged from 57% to 68% and 69% to 77%, respectively. The review also estimated that 87% to 92% participants respond to at least one of these stimulants. However, although MPH and amphetamine can effectively manage ADHD symptoms in most pediatric patients, many patients still fail to respond optimally to either.
Lisdexamfetamine dimesylate (LDX; Vyvanse^®^) is a prodrug stimulant with a novel delivery mechanism, approved in Canada \[[@B12]\] and the United States \[[@B13]\] for the treatment of ADHD in children 6 to 12 years of age, adolescents aged 13 to 17 years, and adults. LDX is a therapeutically inactive molecule. LDX is converted, primarily in the blood, to l-lysine and therapeutically active d-amphetamine \[[@B14]\]. In Canada, the approved dosages range from 20 to 60 mg capsules for once daily oral administration and in the United States from 20 to 70 mg also once daily \[[@B12],[@B13]\].
LDX has been shown to be effective from 1.5 to 13 hours postdose in children \[[@B15]\], and from 2 to 14 hours postdose in adults \[[@B16]\]. In a randomized controlled trial (RCT), LDX was associated with improvements in clinical symptoms of ADHD in children while maintaining a safety profile similar to other stimulant medications \[[@B17]\].
In this post hoc analysis from the RCT, the efficacy of LDX in a subset of children, who had significant ADHD symptoms at study enrollment despite receiving MPH treatment, was evaluated to determine clinical response to LDX therapy in these study participants. Based on previous findings that some patients fail to achieve optimal response to either MPH or amphetamine, children who were previously treated with MPH and continue to have ADHD symptoms may be responsive to amphetamine-based ADHD treatment.
Methods
=======
The methods used in this study for the overall study population have been described previously \[[@B17]\]. This was a multicenter, randomized, double-blind, forced-dose titration, parallel-group study, conducted in accordance with the Guideline for Good Clinical Practice from the World Health Organization and the Declaration of Helsinki and its amendments.
Participants
------------
Biederman et al previously described full inclusion/exclusion criteria \[[@B17]\]. Briefly, children aged 6 to 12 years who met DSM-IV-TR criteria for a primary diagnosis of ADHD \[[@B18]\] and had a ADHD Rating Scale IV (ADHD-RS-IV) \[[@B19],[@B20]\] score of ≥ 28 at baseline after washout were eligible for inclusion, regardless of medication used for ADHD at screening.
Study Design
------------
The study comprised a 1-week screening period; a 1-week washout period of prior psychoactive medications; and 4-weeks of double-blind treatment. During screening, participants received an initial ADHD-RS-IV evaluation. Participants receiving medication for ADHD at enrollment were allowed to continue their medication during the screening evaluation. After screening, the parents/caregivers of eligible participants were instructed to discontinue their prior ADHD medications, if they had not already done so.
Baseline assessments were made after the 1-week washout. Participants were randomized in a 1:1:1:1 ratio (using a block-randomization schedule) to receive double-blind, oral administration of LDX 30 mg/day for 4 weeks, 50 mg/day (30 mg/day for week 1, 50 mg/day for weeks 2 to 4), 70 mg/day (30 mg/day for week 1, 50 mg/day for week 2, 70 mg/day for weeks 3 and 4), or placebo for 4 weeks.
Efficacy Outcome Measures
-------------------------
The primary efficacy outcome was the change in mean ADHD-RS-IV total score from baseline to treatment endpoint, defined as the last postrandomization week for which a score was obtained. ADHD-RS-IV total score assessments were based on investigator interviews with the caregiver and child regarding symptom severity during the preceding week.
Secondary efficacy measures included ADHD-RS-IV total scores at screening, baseline, and endpoint; percent change in ADHD-RS-IV total score; the Conners\' Parent Rating Scale-Revised (CPRS-R: Short Form) \[[@B21]\]; and the investigator-rated Clinical Global Impressions (CGI) scale \[[@B22]\]. The CGI-Severity (CGI-S) assessment was conducted at the baseline visit and the CGI-Improvement (CGI-I) assessment was conducted at subsequent visits.
Efficacy was assessed in the overall efficacy population, all participants who had ADHD-RS-IV scores recorded at baseline and at least one other postrandomization time point.
Post Hoc Efficacy Analyses
--------------------------
This post hoc efficacy analysis assessed treatment effects of LDX and placebo in participants receiving MPH prior to entering the present study, who had available screening data and significant ADHD symptoms prior to discontinuing their MPH regimen. Efficacy was further evaluated according to mean daily MPH dose received (≥ 1 mg/kg vs \< 1 mg/kg) during prior treatment.
Rates of symptomatic remission and clinical response were evaluated throughout the study in participants receiving prior MPH therapy and the efficacy population.
Steele et al \[[@B23]\] suggested that treatment response be considered as an improvement in symptom scores from baseline of 25% to 30%. However, reductions from baseline do not take into account potential differences in baseline severity of disease. Participants with severe symptoms at baseline may be considered responders but still exhibit symptoms. Hence, a clinical response definition that includes a percent reduction in symptoms and a measure of global clinical improvement, such as the CGI-I, may be a better measure of clinical response to treatment. Moreover, other studies have shown that a 1-level change on the CGI-I was consistent with an estimated 10- to 15-point or 25% to 30% change from baseline in ADHD-RS-IV total score \[[@B24]\].
In the primary analysis \[[@B17]\], Biederman et al reported on the ADHD-RS-IV (primary outcome measure) and CGI-I (secondary outcome measure) as continuous measures. In this present analysis, clinical response to LDX treatment was defined as a dual criteria of ≥ 30% reduction in ADHD-RS-IV total score from baseline and a CGI-I score of 1 or 2 at endpoint based on data from previous reports defining response \[[@B23],[@B25]\]; symptomatic remission was defined as ADHD-RS-IV total score of ≤ 18 \[[@B26]\]. Conversely, nonremitters on prior MPH were defined as participants with an ADHD-RS-IV total score \> 18 while receiving MPH prior to entering the study. Number-needed-to-treat (NNT) for 1 participant to achieve a therapeutic clinical response or symptomatic remission at treatment endpoint was calculated to translate the efficacy data into more clinically meaningful terms.
Safety Assessments
------------------
Safety assessments, in enrolled participants who received at least 1 dose of study medication, have been reported previously \[[@B17]\]. Briefly, these included adverse events (AEs), electrocardiograms (ECGs), blood pressure (BP), heart rate, and laboratory assessments. Treatment-emergent AEs (TEAEs) were coded using the *Medical Dictionary for Regulatory Activities*version 7.1 \[[@B27]\]. TEAEs referred to events with onset after the first date of treatment and no later than 3 days following termination of treatment. No separate assessments were performed in nonremitters on prior MPH due to low sample numbers and no reason to expect differences in safety/tolerability in these participants.
Statistical Analyses
--------------------
ADHD-RS-IV total and CPRS-R ADHD Index scores were summarized as mean (standard deviation \[SD\]). Mean change in ADHD-RS-IV total score for the overall population was assessed using 2-way analysis of covariance. Dunnett test for multiple mean comparisons with least-squares adjustment was used to compare change from baseline in the 3 active treatment groups versus placebo. NNT to achieve 1 clinical responder or 1 symptomatic remitter was calculated as the reciprocal of the difference in proportions of clinical responders or symptomatic remitters on active treatment and placebo at treatment endpoint.
Results
=======
Participant Demographics and Disposition
----------------------------------------
In total, 297 children were enrolled at 40 study sites in the United States, of which 7 children discontinued prior to randomization, and 290 were randomized to receive LDX (n = 218) or placebo (n = 72). Of these, 285 had a postrandomization symptom assessment and were included in the efficacy population. Full demographic data for this population have been previously reported \[[@B17]\].
Of the 290 randomized participants, 28 were receiving MPH treatment at screening and 26 of these were classified as nonremitters on prior MPH at the screening visit, prior to randomization (Table [1](#T1){ref-type="table"}). Median age was 9 years and 11/26 (42.3%) female and 15/26 (57.7%) male participants were included. Prior treatment for 19 (73.1%) participants was osmotic, controlled-release MPH (OROS MPH), alone or in combination with another ADHD medication (1 participant in combination with IR dex-MPH \[d-MPH\], 1 with IR mixed amphetamine salts); 2 (7.7%) participants received prior treatment with extended release (ER) MPH; 3 (7.7%) participants received prior treatment with IR MPH; 1 (3.8%) participant was previously treated with sustained release MPH (SR MPH); 1 (3.8%) participant was prior treated with MPH controlled delivery (MPH CD) (Table [1](#T1){ref-type="table"}). Sixteen participants (61.5%) received an average daily dose of ≥ 1 mg/kg MPH, and 10 (38.5%) an average daily dose of \< 1 mg/kg MPH.
######
Baseline Demographic and Clinical Characteristics of Randomized Participants Classified as Nonremitters During Prior MPH Treatment
---------------------------------------------------------------------------------------------------------------------------------------------------------
Participant Age (years) Sex Weight (kg) Medication Total Daily Dose (mg/day) Average Daily Dose (mg/kg) Screening ADHD-RS-IV Total Score
------------- ------------- ----- ------------- ------------- --------------------------- ---------------------------- ----------------------------------
1 6 F 22.68 OROS MPH\* ≥ 30 ≥ 1.0 50
2 7 M 29.48 ER MPH 20 \< 1.0 20
3 7 F 44.45 OROS MPH 36 \< 1.0 28
4 8 M 44.36 OROS MPH 18 \< 1.0 38
5 8 M 26.31 OROS MPH 54 ≥ 1.0 38
6 8 M 33.57 OROS MPH;\ 54;\ ≥ 1.0 50
IR dMPH 2.5
7 8 M 41.28 OROS MPH 18 \< 1.0 43
8 8 F 43.68 IR MPH 30 \< 1.0 50
9 9 M 29.03 ER MPH 20 \< 1.0 29
10 9 M 26.76 OROS MPH 54 ≥ 1.0 39
11 9 M 31.75 MPH CD 40 ≥ 1.0 29
12 9 M 26.31 OROS MPH 27 ≥ 1.0 45
13 9 F 28.12 SR MPH 20 \< 1.0 40
14 9 F 25.18 IR MPH\* ≥ 50 ≥ 1.0 34
15 9 F 24.90 OROS MPH 27 ≥ 1.0 20
16 10 M 39.01 OROS MPH 54 ≥ 1.0 23
17 10 F 43.68 OROS MPH^†^ 36 \< 1.0 22
18 10 F 27.67 OROS MPH 54 ≥ 1.0 37
19 10 F 29.03 IR MPH 50 ≥ 1.0 44
20 11 M 45.36 OROS MPH 72 ≥ 1.0 35
21 11 M 45.36 OROS MPH 36 \< 1.0 41
22 12 M 39.46 OROS MPH 54 ≥ 1.0 45
23 12 M 34.02 OROS MPH 18 \< 1.0 45
24 12 M 33.57 OROS MPH 54 ≥ 1.0 44
25 12 F 34.02 OROS MPH 36 ≥ 1.0 51
26 12 F 26.76 OROS MPH 54 ≥ 1.0 25
---------------------------------------------------------------------------------------------------------------------------------------------------------
ADHD-RS-IV = Attention-Deficit/Hyperactivity Disorder Rating Scale IV; dMPH = dexmethylphenidate; ER = extended-release; IR = immediate-release; MPH = methylphenidate; CD = controlled delivery; OROS = osmotic-release oral system; SR = sustained-release.
\*Exact dose of treatment for these participants could not be determined; ^†^Participant was also receiving 40 mg/d of IR mixed amphetamine salts although this was not included in the calculation of MPH dose.
Changes in ADHD-RS-IV Total Scores
----------------------------------
Mean (SD) screening, baseline, and endpoint ADHD-RS-IV total scores for nonremitters during prior MPH treatment, nonremitters stratified according to prior MPH dosage received, and overall efficacy population are shown in Figure [1](#F1){ref-type="fig"}.
{#F1}
The mean (SD) change in ADHD-RS-IV total score from baseline with LDX treatment was -24.0 (12.56) (Figure [2](#F2){ref-type="fig"}), corresponding to a mean (SD) percentage reduction of 57 (29.9%) in the 19 nonremitters on prior MPH treatment. The mean (95% confidence interval \[CI\]) placebo-adjusted ADHD-RS-IV total score reduction for this group was -17.6 (-29.65, -5.49; *P*= .0063).
{#F2}
Changes in ADHD-RS-IV and CGI-I Scores: Remitters and Responders
----------------------------------------------------------------
Of the 26 nonremitters on prior MPH at screening, 12 (63.2%) participants receiving LDX and 1 (14.3%) receiving placebo were classified as remitters during the study (Figure [3](#F3){ref-type="fig"}). Similar patterns of symptomatic remission with LDX treatment were observed in the overall efficacy population. As well, patterns of symptomatic remission were similar with placebo treatment in the overall efficacy population and in nonremitters on prior MPH. The NNT (95% CI) to achieve symptomatic remission with LDX at treatment endpoint was 2.0 (1.21, 6.63) in nonremitters and 2.1 (1.74, 2.72) in the overall study population.
{#F3}
Of nonremitters on prior MPH, clinical response was achieved in 15 (78.9%) treated with LDX and 3 (42.9%) treated with placebo, respectively. In the overall efficacy population, 169 (79.3%) treated with LDX and 21 (29.2%) treated with placebo achieved clinical response (Figure [4](#F4){ref-type="fig"}). Of the 169 LDX clinical responders, 54 (32.0%) received 30 mg/d LDX, 55 (32.5%) received 50 mg/d LDX, and 60 (35.5%) received 70 mg/d LDX. NNT (95% CI) to achieve clinical response with LDX at treatment endpoint was 2.0 (1.21, 6.63) in nonremitters on prior MPH, versus 1.8 (1.51, 2.22) in the overall population.
{#F4}
Changes in CPRS-R ADHD Index Scores
-----------------------------------
Mean (SD) morning, afternoon, and evening CPRS-R ADHD index scores at baseline and endpoint in nonremitters on prior MPH are shown in Figure [5](#F5){ref-type="fig"}. The mean changes from baseline morning, afternoon, and evening CPRS-R ADHD index scores were -14.7 (10.90), -12.2 (12.89), and -13.4 (11.69) for the LDX groups, respectively, and -1.3 (14.92), -0.1 (9.01), and 0.4 (11.25) for the placebo group, respectively. These data were similar to the CPRS-R ADHD index scores observed in the overall population \[[@B17]\].
{#F5}
Safety and Tolerability
-----------------------
Full safety analyses have been reported previously \[[@B17]\]. In the safety population, 196/290 (68%) participants reported one or more TEAEs; 21/290 (7.2%) discontinued due to TEAE. TEAEs with an incidence ≥ 5% in the combined LDX group were decreased appetite, insomnia, headache, upper abdominal pain, irritability, weight loss, vomiting, nausea, dizziness, and nasopharyngitis and, in the placebo group, were headache, cough, nasal congestion, nasopharyngitis, and upper abdominal pain. No serious AEs were observed during the study. More than 95% of TEAEs were mild or moderate in intensity and most began during the first week of treatment and abated over time \[[@B17]\]. Mean (SE) change from baseline at endpoint for pulse (bpm) ranged from 0.3 (1.20) to 4.1 (1.17) in all LDX groups and was -0.7 (1.17) in the placebo group. The systolic BP change for all LDX groups ranged from 0.4 (1.08) to 2.6 (1.05) mm Hg and for placebo was 1.3 (1.05) mm Hg. For diastolic BP the change ranged from 0.6 (0.93) to 2.3 (0.91) mm Hg for all LDX groups and was 0.6 (0.91) mm Hg for the placebo group. LDX treatment was not associated with any significant changes in mean BP, ECG parameters, and laboratory values.
Discussion
==========
In this post hoc analysis, LDX showed efficacy when given to children with significant clinical ADHD symptoms despite prior MPH treatment. Efficacy outcomes were similar to the results of the overall population assessed in the clinical trial.
Among participants previously treated with MPH, more than half were receiving doses (average daily dose ≥ 1 mg/kg) considered generally effective according to the regimens administered in randomized, controlled trials \[[@B28],[@B29]\]. Conversely, just under half may have received suboptimal doses. Moreover, none of these measures differed from those observed in the overall study population. Although this study was not powered to detect differences between the treatment groups, the percentage of clinical responders in the overall study group was comparable regardless of LDX dose received. Similarly, no apparent differences occurred between the NNTs to achieve clinical response or symptomatic remission for the overall efficacy population and nonremitters on prior MPH. The NNT values calculated are comparable or superior to those reported elsewhere in the literature for symptomatic remission and clinical response to MPH and atomoxetine, which range from approximately 1.9 to 5.3 depending on formulation and types of raters \[[@B30]\].
Differential responses to MPH and amphetamine may explain a successful clinical response to LDX in participants who had significant ADHD symptoms despite prior MPH therapy. In 2 separate crossover studies \[[@B31],[@B32]\] comparing the efficacy of MPH and dextroamphetamine, most children with ADHD who did not respond to 1 stimulant responded to the other. A bimodal pattern of clinical response to atomoxetine has been described, with no obvious demographic or clinical predictors of clinical response \[[@B33]\].
Clinical trial design may have contributed to the observed clinical response to LDX treatment in nonremitters on prior MPH. LDX treatment was administered in a forced-dose titration, while prior MPH therapy was provided according to community standards and included potential suboptimal dosing. Use of different definitions of therapeutic response may have altered the rates observed.
This post hoc analysis has limitations. The classification of participants as nonremitters on prior MPH considers only the ADHD-RS-IV total score at screening and may not reflect the participants\' overall clinical response to MPH. It should be noted that switching from MPH formulations to LDX was done as part of the study protocol and not purely as a clinical practice decision.
The 4-week study duration limits the ability to extrapolate the findings to the long-term treatment generally required in managing ADHD. This study was not prospectively designed or powered to detect differences between the treatment groups. A prospective study would be required to confirm these preliminary findings.
Conclusions
===========
In this post hoc analysis of children who had significant clinical ADHD symptoms despite previous MPH treatment, LDX demonstrated efficacy and clinical response in the subpopulation assessed. Efficacy outcomes in this population were similar to those in the overall study population.
List of Abbreviations
=====================
ADHD: attention-deficit/hyperactivity disorder; ADHD-RS-IV: ADHD Rating Scale IV; AE: adverse event; BP: blood pressure; CADDRA: Canadian ADHD Resource Alliance; CD: controlled delivery; CGI-S: Clinical Global Impressions-Severity; CGI-I: Clinical Global Impressions-Improvement; CI: confidence interval; CPRS-R: Short Form: Conners\' Parent Rating Scale-Revised; dMPH: dexmethylphenidate; DSM-IV-TR*: Diagnostic and Statistical Manual of Mental Disorders*, *Fourth Edition, Text Revision*; ECG: electrocardiogram; ER: extended-release; IR: immediate-release; LDX: lisdexamfetamine dimesylate; MAS: mixed amphetamine salts; MPH: methylphenidate; NNT: number-needed-to-treat; OROS: osmotic release oral system; RCT: randomized controlled trial; SR: sustained-release; TEAE: treatment-emergent AE
Competing interests
===================
Dr Jain or Saundra Jain receives or has received grant research support from Abbott, Addrenex, Aspect, Forest, Lilly, and Pfizer; served as a consultant for Addrenex, Impax, Lilly, and Shire; served on a speaker\'s bureau for Cyberonics, GlaxoSmithKline, Jazz, Pfizer, Shire, and Takeda; received honorarium from Cyberonics, Forest, Jazz, Lilly, Pfizer, Roche, Shire, and Takeda. Dr Babcock is formerly an employee of Shire and holds stock and/or stock options in Shire. Dr Burtea is an employee of Shire Canada Inc. and holds stock and/or stock options in Shire Canada Inc. Dr Dirks is an employee of Shire and holds stock and/or stock options in Johnson & Johnson and Shire. Mr Adeyi is an employee of Shire and holds stock and/or stock options in Shire. Dr Scheckner is an employee of Shire and holds stock and/or stock options in Shire. Dr Lasser is an employee of Shire and holds stock and/or stock options in Shire.
Authors\' contributions
=======================
RJwas an investigator on the parent study and participated in data acquisition, analysis, interpretation, and presentation. RJ was fully involved in drafting the manuscript and revising the intellectual content of this manuscript. He has given final approval of this version. TBabcockwas the associate director, Scientific Publications, Clinical Development, and Medical Affairs for this study, and made substantial contributions to the analysis and interpretation of the data. He was deeply involved in drafting the manuscript and revising the intellectual content. He has given final approval of this version. TBurteawas the medical director, Global Clinical Development and Medical Affairs for this study and made substantial contributions to the analysis, and interpretation of the data. He was deeply involved in drafting the manuscript and revising the intellectual content. He has given final approval of this version. BDwas the director, Clinical Development and Medical Affairs for this study, and made substantial contributions to the analysis and interpretation of the data. He was deeply involved in drafting the manuscript and revising the intellectual content. He has given final approval of this version. BAwas a statistician involved in all post hoc data analysis, interpretation, and presentation. Statistician BA was fully involved in drafting and revising the intellectual content of this manuscript. Statistician BA has given final approval to this version. BSwas the associate director, Scientific Publications, Clinical Development, and Medical Affairs for this study, and made substantial contributions to the analysis and interpretation of the data. He was deeply involved in drafting the manuscript and revising the intellectual content. He has given final approval of this version. RLwas the senior director, Clinical Development and Medical Affairs for this study, and made substantial contributions to the analysis and interpretation of the data. He was deeply involved in drafting the manuscript and revising the intellectual content. He has given final approval of this version.
Acknowledgements
================
Clinical research was funded by the sponsor, Shire Canada Inc. Under the direction of the authors, Kira Belkin and William Perlman, employees of Excerpta Medica, and Huda Ismail Abdullah, PhD, and Michael Pucci, PhD, employees of SCI Scientific Communications & Information (SCI), provided writing assistance for this publication. Editorial assistance in formatting, proofreading, copy editing, and fact checking was also provided by Excerpta Medica and SCI. Robert Morgan from Shire Canada Inc. also reviewed and edited the manuscript for scientific accuracy. Shire Canada Inc. provided funding to Excerpta Medica and SCI for support in writing and editing this manuscript. Although the sponsor was involved in the design, collection, analysis, interpretation, and fact checking of information, the content of this manuscript, the ultimate interpretation, and the decision to submit it for publication in *Child and Adolescent Psychiatry and Mental Health*was made by the authors independently.
|
{
"pile_set_name": "PubMed Central"
}
|
Yeah, once again, you prove just how amazing you are... don't ever stop being you, never give in to the world, always fight and always be our Amarynceus!! We love you!!
|
{
"pile_set_name": "OpenWebText2"
}
|
#ifndef SHAREMENUTOOLBAR_H
#define SHAREMENUTOOLBAR_H
#include <QtWidgets>
class ShareToolbar : public QToolBar {
Q_OBJECT
public:
ShareToolbar(QWidget *parent = nullptr);
public slots:
void setLeftMargin(int value);
};
#endif // SHAREMENUTOOLBAR_H
|
{
"pile_set_name": "Github"
}
|
Download Angry Bird Space V1.0.2 for PC | Download Game Terbaru Gratis. Angry Bird Space V1.0.2 for PC is the one of Angry Bird Game for PC that very good and you should to Download and try it. You can get the game by Download torrent, mediafire, megaupload, mediahide, torrent, cnet, filesonic, etc. Ok Guys, lets check the Game here:
Release name: Angry_Birds_Space_v1.0.2_CRACKED-P2P
Size: 34.70 MB
Release Date: 24.03.2012
Language: ENG
Release Description:
The chase is on! After a giant claw kidnaps their eggs, the Angry Birds chase it into a wormhole and find themselves floating in a strange new galaxy – surrounded by space pigs! Luckily the Angry Birds have super powers of their own… Angry Birds Space introduces you to new adventures on planets and in zero gravity, resulting in spectacular gameplay ranging from slow-motion puzzles to lightspeed destruction. With brand new birds, brand new superpowers, and a whole galaxy to explore, the sky is no longer the limit! Also featuring the DANGER ZONE, the most difficult Angry Birds levels ever! Can you master the Danger Zone?
|
{
"pile_set_name": "Pile-CC"
}
|
Conversion of antinuclear antibody specificity as a marker of deterioration of cutaneous lupus erythematosus into lupus nephritis.
A 34-year-old male systemic lupus erythematosus patient (SLE) with cutaneous vasculitis developed renal failure after switching anti-nuclear antibody (ANA) specificity. He developed cutaneous lupus erythematosus with homogeneous and speckled type ANA and a high titer of anti-DNA antibody without renal involvement at 21 years of age. After developing lupus nephritis at the age of 27, the original ANA disappeared gradually. Two years later, a discrete speckled type ANA titer elevated abruptly to as high as 1:640 with low complementemia and without DNA antibody. Within five years, he suffered renal failure. This case of SLE suggests a direct correlation with ANA pattern and organ involvement.
|
{
"pile_set_name": "PubMed Abstracts"
}
|
Diabetes is one of the fastest growing diseases today. The World Health Organization estimates that 177 million people worldwide currently have diabetes and this number is projected to increase to more than 370 million people by the year 2030. The costs associated with diabetes, including premature death, pain and suffering, and increased financial burdens. These costs are directly related to the medical complications associated with chronic hyperglycemia. Early detection and maintaining a tight glycemic control are paramount to controlling the costs of the diabetic epidemic.
The cornerstone of tight glycemic control is frequent glucose monitoring, where blood glucose concentrations are measured to help administer proper levels of insulin and maintain euglycemic conditions. To this end, glucose sensing technology has advanced considerably in recent years to provide tools for home glucose monitoring and establishing opportunities for tight glycemic control. The current conventional determination of blood glucose is a routine invasive procedure typically performed several times a day. In general, this procedure involves the taking of a small blood sample and evaluating the level of glucose in the sample. Common instruments used for this use the enzyme glucose oxidase to convert glucose and oxygen to gluconic acid and hydrogen peroxide. The level of hydrogen peroxide is then measured by spectroscopic or electrochemical means which is reflective of the starting glucose concentration.
While these daily measurements provide a diabetic patient with the ability to self-monitor and thus better control blood glucose levels, they are not without drawbacks. In particular, the taking of blood samples several times daily can be very painful and expose the patient to elevated risks of infection. Moreover, these methods are not suitable for providing continuous blood glucose measurements. Thus, for example, during the night, a patient must either be awakened periodically for testing or else run the risk that glucose levels will drop dangerously low while they sleep.
Non-invasive optical sensing of an analyte, such as glucose, has been proposed as an approach for frequent and painless measurement of glucose in diabetics. However, to date, all reported attempts to measure glucose non-invasively have involved collecting spectra from a human and then using a classical statistical multivariate calibration technique to correlate variations in the spectral information to blood glucose concentrations. These statistical techniques rely on regressions to statistically correlate spectral variances to an artificially assigned glucose concentration. Thus, these measurements are not necessarily based on actual analyte specific spectral features. Further, these statistical methods fail to provide direct evidence that the assigned concentration predictions from the multivariate calibration models are actually based on glucose specific spectral information. Moreover, in some cases the in vivo spectral signature for a physiological analyte can be smaller than many weakly or partially correlated spectral variations, making the use of the conventional statistical methods very difficult.
Therefore, in view of the foregoing, there exists a need for an in vivo calibration method that can identify analyte specific spectral information. Further, there is also a need for a non-invasive method of measuring the concentration of an analyte in a test subject. Moreover, there is also a need for a method for evaluating the analytical significance of the classical statistical multivariate calibration models.
|
{
"pile_set_name": "USPTO Backgrounds"
}
|
Dominic McGuire
Dominic Rashad McGuire (born October 20, 1985) is an American professional basketball player who plays for Real Estelí Baloncesto.
High school career
McGuire played high school prep basketball at Lincoln High School in San Diego.
College career
After high school, McGuire subsequently attended the University of California, Berkeley, where he played college basketball with the Cal Golden Bears. He transferred to California State University, Fresno to play with the Fresno State Bulldogs, after his sophomore year of college, when he was the fourth-leading scorer for the Cal Golden Bears the during the 2004–05 season.
Professional career
Washington Wizards (2007–2010)
McGuire was selected in the second round (47th overall) of the 2007 NBA Draft out of Fresno State by the Washington Wizards. On August 13, 2007, McGuire signed a 3-year deal with the Washington Wizards. On April 16, 2008, McGuire got his first start in his NBA career, against the Orlando Magic.
Sacramento Kings (2010)
On February 18, 2010, he was traded to the Sacramento Kings for a protected 2010 2nd round draft pick.
Charlotte Bobcats (2010–2011)
On July 22, 2010, he signed with the Charlotte Bobcats. On February 24, 2011, McGuire was waived by the Bobcats following a trade between the Bobcats and Portland Trail Blazers. However, he was re-signed on March 3, 2011.
Golden State Warriors (2011–2012)
On December 15, 2011, McGuire signed with the Golden State Warriors.
Toronto Raptors (2012)
On September 12, 2012, he signed with the Toronto Raptors. He was waived on November 30, 2012, when the team signed Mickael Pietrus.
New Orleans Hornets (2012–2013)
On December 16, 2012, McGuire signed with the New Orleans Hornets. He was waived by the Hornets on January 4, 2013.
Indiana Pacers (2013)
On January 7, 2013, McGuire signed with the Indiana Pacers on a 10-day contract. He was signed to another 10-day contract on January 17, 2013.
Utah Jazz (2013)
On September 26, 2013, he signed with the Utah Jazz. He was later waived by the Jazz on October 26.
Santa Cruz Warriors (2014)
In January 2014, he was acquired by the NBA D-League's Santa Cruz Warriors. In March 2014, he was traded to the NBA D-League's Tulsa 66ers.
Gigantes de Guayana (2014)
In April 2014, he signed with Gigantes de Guayana of Venezuela for the rest of the 2014 LPB season.
Hapoel Eilat (2014)
On August 27, 2014, he signed to play in Israel with Hapoel Eilat of the Israeli Basketball Premier League.
Hapoel Holon (2014–2015)
On November 6, 2014, he left Eilat and signed with another Israeli club, Hapoel Holon for the rest of the season.
Shenzhen Leopards (2015–2016)
In August 2015, McGuire signed with the Shenzhen Leopards of China for the 2015–16 CBA season.
Leones de Ponce (2016)
On February 26, 2016, he signed with the Leones de Ponce of Puerto Rico for the 2016 BSN season. On April 14, 2016, he parted ways with Leones de Ponce.
Club Malvín (2018)
On April 19, 2018, McGuire was reported to have signed with Club Malvín of the Liga Uruguaya de Basketball.
Aguacateros de Michoacán (2018–present)
On December 2018, McGuire joined Mexican team Aguacateros de Michoacán of the Liga Nacional de Baloncesto Profesional.
NBA career statistics
Regular season
|-
| align="left" |
| align="left" | Washington
| 70 || 1 || 9.9 || .379 || .167 || .438 || 2.0 || .6 || .3 || .4 || 1.3
|-
| align="left" |
| align="left" | Washington
| 79 || 57 || 26.2 || .432 || .500 || .725 || 5.4 || 2.5 || .8 || .9 || 4.5
|-
| align="left" |
| align="left" | Washington
| 41 || 0 || 5.9 || .375 || .000 || .000 || 1.5 || .2 || .1 || .1 || .7
|-
| align="left" |
| align="left" | Sacramento
| 10 || 2 || 6.7 || .333 || .000 || .000 || 1.8 || .3 || .1 || .1 || .8
|-
| align="left" |
| align="left" | Charlotte
| 52 || 8 || 14.6 || .396 || .000 || .769 || 3.8 || .8 || .2 || .6 || 3.3
|-
| align="left" |
| align="left" | Golden State
| 64 || 6 || 17.6 || .448 || .000 || .736 || 3.8 || 1.7 || .7 || .6 || 3.5
|-
| align="left" |
| align="left" | Toronto
| 15 || 9 || 15.3 || .469 || .000 || .333 || 3.2 || .7 || .3 || .5 || 2.1
|-
| align="left" |
| align="left" | New Orleans
| 9 || 0 || 16.1 || .429 || .000 || .500 || 3.1 || 1.0 || .9 || .3 || 2.1
|-
| align="left" |
| align="left" | Indiana
| 2 || 1 || 6.0 || .000 || .000 || .000 || 1.0 || .5 || .0 || .0 || .0
|- class="sortbottom"
| style="text-align:center;" colspan="2"| Career
| 342 || 84 || 15.6 || .419 || .188 || .658 || 3.4 || 1.2 || .5 || .5 || 2.7
Playoffs
|-
| align="left" | 2008
| align="left" | Washington
| 3 || 0 || 5.0 || .000 || .000 || .500 || 1.0 || .3 || .3 || .3 || .3
|- class="sortbottom"
| style="text-align:center;" colspan="2"| Career
| 3 || 0 || 5.0 || .000 || .000 || .500 || 1.0 || .3 || .3 || .3 || .3
Career highs
Points: 16 @ Philadelphia 01/30/09
Rebounds: 17 @ Atlanta 12/17/10
Assists: 9 vs. Atlanta 03/02/09
Steals: 4 vs. Phoenix 01/26/09
Blocks: 4 3 times
References
External links
Career statistics and player information from NBA.com
Eurobasket profile
Category:1985 births
Category:Living people
Category:African-American basketball players
Category:American expatriate basketball people in Canada
Category:American expatriate basketball people in China
Category:American expatriate basketball people in Israel
Category:American expatriate basketball people in Mexico
Category:American expatriate basketball people in Uruguay
Category:American expatriate basketball people in Venezuela
Category:American men's basketball players
Category:Basketball players from California
Category:Big3 players
Category:California Golden Bears men's basketball players
Category:Charlotte Bobcats players
Category:Club Malvín basketball players
Category:Fresno State Bulldogs men's basketball players
Category:Golden State Warriors players
Category:Hapoel Eilat basketball players
Category:Hapoel Holon players
Category:Leones de Ponce basketball players
Category:Power forwards (basketball)
Category:Sacramento Kings players
Category:Santa Cruz Warriors players
Category:Shenzhen Aviators players
Category:Small forwards
Category:Sportspeople from San Diego
Category:Toronto Raptors players
Category:Washington Wizards draft picks
Category:Washington Wizards players
Category:Real Estelí Baloncesto players
|
{
"pile_set_name": "Wikipedia (en)"
}
|
The president's 2013 budget, released today, asks for modest increases for some federal science agencies but trims funding to NASA. The request takes a deep bite out of Mars and outer-planet science exploration in particular.
NASA's funding would fall to the lowest level in four years, with a total budget of $17.71 billion. The president's request projects a flat budget through 2017, with no growth to even account for inflation.
"We are having to make tough decisions because these are tough economic times," said NASA administrator Charles Bolden during a press conference Feb. 13.
As expected, planetary science – in particular Mars exploration and outer-planets missions – is the biggest loser, getting a $309 million decrease compared to last year. This means NASA will not be able to maintain previous commitments to the European Space Agency for dual Mars missions in 2016 and 2018, Bolden confirmed at the conference.
The hit to outer-planet exploration means a lack of funding for any new mission to study the moons of Jupiter or a Uranus orbiter, two projects that received high priority in last year's planetary science decadal survey. The reduction might also affect ongoing missions such as the Cassini spacecraft that is currently exploring Saturn and its moons, though this will depend on the outcome of NASA's senior reviews later this year.
The budget does include some winners. Manned exploration would get a boost of $200 million. This includes $2.8 billion for a new heavy-lift rocket system, which continues much of the work from the canceled Bush-era Constellation program that would take astronauts beyond low-Earth orbit.
The long-delayed and over-budget James Webb Space Telescope would get $627.6 million, which is a more than $100 million increase over last year. This mission will require continued funding until its launch date, currently pegged at October 2018.
Other scientific agencies are also getting modest increases. The National Science Foundation would be given a 5 percent increase, receiving $7.4 million. While Obama promised to double the NSF's budget back in 2009, the difficult fiscal environment doesn’t seem to make this possible. The U.S. Geological Survey could also be getting a $34.5 million boost compared to last year, bringing its funding to $1.1 billion.
The president's budget request is just a taste of what may come. Congress has to agree on what the actual federal budget looks like, and legislators and the White House don't see eye to eye on many issues.
One major area of contention is the requested $830 million for private spaceflight companies. Many Congress members, particularly those whose districts include traditional spaceflight companies such as Boeing, have scoffed at previous private spaceflight requests, eventually providing around half of the requested amount.
But the new private companies are looking to prove their worth. SpaceX will launch its Dragon capsule to dock with the International Space Station this year.
The budget request also makes no mention of the failure of last year's bipartisan congressional subcommittee that was meant to find $1.2 trillion in deficit reductions. Members of this committee were unable to compromise, triggering an across-the-board cut to all government agencies. Congress will have to figure out what to do with this legislative requirement before starting deliberations on the final 2013 budget.
|
{
"pile_set_name": "OpenWebText2"
}
|
The game that pioneered the 3D fighting genre is back with Virtua Fighter 5, the latest instalment in the popular series, currently under development for the Microsoft Xbox 360. Virtua Fighter 5 will elevate the arcade fighting genre to all new heights as the game promises to take true advantage of the capabilities of the next generation hardware. Virtua Fighter 5 raises the bar for console fighting games including all the features fans know and love plus enhanced gameplay mechanics, additional characters and new fighting styles, as well as newly redesigned 3D environments.
Virtua Fighter 5 will deliver fast-paced, adrenaline-pumping action as players head into battle, taking on a host of popular characters. The game will introduce two new dynamic characters, El Blaze and Eileen, complete with new fighting techniques and from completely different backgrounds. Play as one of the 17 default characters in the game or customize a character. Players will be able to modify their characters by selecting from four base costumes and then decorate them by attaching a wide range of unlockable and earnable items. Players will not only achieve victory by defeating highly-skilled opponents, but also by competing for prizes and earning in-game money allowing them to buy many items at an in-game shop. Further building upon the depth of the series, players will now be able to move around their opponent using an "Offensive Move" technique, adding a new strategic element to their battles.
Key Features:
Two New Characters, El Blaze and Eileen, round out the cast of 17 dynamic characters. El Blaze is a Mexican wrestling champion that uses the Lucha Libre fighting style, and Eileen, originally from China, uses "Kou-Ken" a Monkey Kung Fu style which she learned from her grandfather. Reinvented Quest Mode where players can compet
CERTAIN CONTENT THAT APPEARS ON THIS SITE COMES FROM AMAZON SERVICES LLC. THIS CONTENT IS PROVIDED ‘AS IS’ AND IS SUBJECT TO CHANGE OR REMOVAL AT ANY TIME.
|
{
"pile_set_name": "Pile-CC"
}
|
The FBI is sitting on more than 641m photos of people’s faces
|
{
"pile_set_name": "OpenWebText2"
}
|
---
author:
- 'Santosh Aditham, and Nagarajan Ranganathan, '
title: |
A System Architecture for the Detection of\
Insider Attacks in Big Data Systems
---
data solutions are widely adopted across various government and enterprise domains such as software, finance, retail and healthcare. Big data applications are pioneering in the field of advanced data analytics and have a projected market of approximately 50 billion dollars by 2018. The most frequent use-cases of big data are information retrieval from complex, unstructured data; and real time data analysis [@IDC]. Along with its rapid market growth, the big data trend also has its share of challenges and risks. In an era where extracting information from data is sanctioned to all, users are understandably more skeptical to let providers host their data away from them. This, along with the recent increase in the number of cyber attacks, boosted the importance for security. Yet, the losses due to boundless security holes in existing systems seem to overshadow the investments towards increasing their security. Hence, there is an immediate need to address architectural loopholes in order to provide better security. For instance, current big data security platforms focus on providing fine-grained security through extensive analysis of stored data. But such models indirectly facilitate the abuse of user data in the hands of the provider.
Insider attacks are becoming more common and are considered the toughest attacks to detect [@Vormetric]. There does not exist much in the literature on solutions for insider attacks in general [@Salem]. Though privacy and security are touted to be important problems in the big data world, the solutions concentrate only on leveraging big data systems for efficient security in other domains. To the best of our knowledge, there is no robust solution for detecting or preventing insider threats within big data infrastructures. For example, security mechanisms of popular big data systems such as Hadoop [@Hadoop] and Spark [@Spark] include third-party applications such as Kerberos [@Kerberos], access control lists (ACL), log monitoring and data encryption (to some extent). But for an insider, especially a traitor, circumventing these mechanisms is not difficult [@Aditham]. It is crucial to address the problem of insider attacks in big data systems for three main reasons: (a) traitor within the provider’s organization will be able to circumvent the security system in place (b) sensitivity of customer information stored in the system is increasing by day; and (c) there is no consensus or widespread agreement on well-defined security standards in the big data community.
Recently, two unauthorized backdoors were discovered in Juniper Networks firewalls that might have given attackers access to highly classified information. Some important facts about this particular hack are: (a) it comes at the cost of compromising national security (b) it shows that even a major network security company is vulnerable to attacks (c) in spite of the high stakes and vast resources, it is believed that these backdoors were left undiscovered for almost 3 years; and (d) it was reported that the attackers could have deleted the security logs [@Swati]. This is one of the many examples to show that the efficiency of common attack prevention techniques, such as identity management, ACLs and data encryption, is necessary but sufficient to prevent attacks. As per OpenSOC, in 60% of breaches data gets stolen within hours of the breach and 54% of breaches are not discovered for months [@Sirota]. This indicates that infrastructures need to have efficient *attack detection* techniques along with strong *attack prevention* techniques for robust security.
In the big data world, it is considered that moving computation to where the data resides is better than the traditional approach of moving data for computation. The main features of big data infrastructures are fast data processing, high scalability, high availability and fault-tolerance. Availability and fault-tolerance of big data systems comes from intelligent replication of data. This implies SIMD style, parallel execution of the same program at multiple locations. When a program is scheduled for execution on the big data cluster, it runs as an individual process on every data node that hosts a copy of the program data. The replication of data on various nodes in the big data system can be utilized in providing security. Security for a computing system can be implemented at hardware and software level. Given the advantage of isolation that can be achieved at hardware level security, we propose delegating security to special purpose hardware, such as TPM [@TPM] and TXT [@TXT] chips, that reside on the nodes of the big data cluster. Such an infrastructure will have the advantages of (a) performing security analysis remotely (b) reducing the overhead on main processor by delegating security, and (c) significantly decrease the cost of data transfer while providing efficient security techniques such as isolated vulnerability scanning through program profiling.
In this paper, we propose a system architecture for attack detection in big data systems that can efficiently detect insider attacks. Our proposed system uses a two step algorithm for attack detection. First, *program profiling* is performed by individual nodes of the big data cluster on the processes they execute. In this step, process binaries of scheduled processes are disassembled and analyzed to generate control instruction sequences (CIS). These sequences are then hashed, encrypted and shared among data nodes that host the same data i.e. primary and replica nodes. Next, *consensus* among data nodes is achieved regarding the possibility of a process being attacked. This step involves two phases: hash matching and information sharing. Upon receiving encrypted messages from primary nodes, the replica nodes apply sequential, on-demand string matching between the locally generated hash and the received hash. Next, the result of this comparison is shared with the primary node. Depending on the results received, the primary data node notifies the master node to take necessary recovery measures. All communications among data nodes are performed using a *secure communication protocol* that is based on public-private key encryption. Our main contributions are as follows:
- We propose a novel extrinsic workflow for security in big data systems using control instruction sequences (CIS), hash matching and encrypted communication.
- We suggest using a one-shot program profiling technique that builds instruction-level CIS from the native code of scheduled processes.
- We endorse the idea of having security as an independent module in big data systems by designing a system architecture for detecting insider attacks in big data systems.
The paper is organized as follows: Section \[sec:background\] gives a primer on insider attacks in general purpose computing. This section also discusses the current security methods in big data and some related works. Section \[sec:model\] describes our attack model. The proposed system which includes the two step attack detection algorithm and the secure communication protocol are explained in Section \[sec:proposed\]. A model of the proposed system with a list of required components is also given in this section. Section \[sec:experiments\] shows the impact and usefulness of the proposed security system architecture by conducting real-world experiments on Hadoop and Spark clusters. Finally, section \[sec:conclusion\] draws the conclusion and outlines future work.
Background & Related Work {#sec:background}
=========================
This section gives a primer on insider attacks and their solutions in general purpose computing and discusses the current security mechanisms available in the big data world. Also, the various related works are briefly described here.
![Entities and Relationships in Insider Attacks[]{data-label="fig_ia"}](insiderattacks.pdf){width="50.00000%"}
Insider Attacks
---------------
Though security in general computing has been extensively studied and implemented over the years, computers are still vulnerable to attacks. Software based attacks that typically target a computer network or system, called cyberattacks, are growing in their frequency and impact. The plot for any type of software attack involves exploitation of a piece of code that runs on a computer. It is inherent to this perspective about a cyberattack that security can be provided at two levels: (a) by the software that is used to compile and execute the program; and (b) by the hardware that runs the program. Providing security at software level gives more context and information about the target programs that are being protected. But this comes with the risk of the security software itself being compromised. On the other hand, having security at hardware level gives more isolation to the process of analyzing and securing programs though it becomes difficult to give detailed context about the programs and the infrastructures running them. In any case, the toughest software attacks to counter are the ones whose genesis is intentional and are performed by those who have a good understanding of the underlying system.
Based on our literature review, we have identified four major questions that can guide towards better handling of insider attacks: (a) who can perform these attacks? (b) what gets affected? (c) how to detect these attacks? and (d) how to prevent them from happening? Figure \[fig\_ia\] gives a list of entities to consider when dealing with insider attacks. The figure also shows the four questions, from above, as relationships among the entities. Insider attacks can be performed by (a) *traitors* who are legally a part of the system but want to misuse the access privileges given to them; (b) *masqueraders* who get access to the system by stealing identities of those who have legitimate access. Insider attacks can affect the proper functionality of a program or corrupt the data used by the programs. Profiling and trapping are two most common ways to detect insider attacks [@Salem; @Schultz]. Profiling can be performed (a) at the program level [@Anup] and at the user level [@Lunt]. Traps can be set in the programs or in the network to force the attacker into performing certain actions that help towards exposing the attack [@Spitzner]. The biggest concern with these insider attack detection methods is the possibility of losing valuable data. Hence, insider attack prevention mechanisms such as identity management [@Froomkin; @Khalil2], access control lists [@Ravi; @Kerberos], data encryption [@Don; @Goyal] etc must be employed at the same time.
In this work, we are more interested in Control Flow Integrity (CFI) [@Ligatti; @Ligatti2] which is another popular and effective technique for attack prevention which enforces the execution of a program to follow a path that belongs to the program’s control flow graph. The set of possible paths are determined ahead of time using static CFG [@Ligatti; @Ligatti2]. A coarse-grained or fine-grained version of CFI can be used for program profiling. But the problem with any such profiling techniques is the overhead incurred in conducting them, even more if performed remotely. Though such limitations of this approach have been identified [@Jujutsu], it is accepted as a strong and stable security enforcing mechanism. There are a plethora of CFG-based code similarity algorithms [@Chan]. But such CFG similarity check methods are complex, expensive, have no defined standards. Most CFG similarity algorithms rely on some simplification techniques such as fingerprints, edit distance, comparison only with known graphs in a database etc. Also, the impact of CFG similarity analysis differs a lot depending on when and how the CFG is generated for a program. These complexities and uncertainties led to a new set of control flow analysis techniques that avoid translating the program code to a formal model. For example, insider attack detection based on symbolic execution and model-checking of assembly code was proposed in [@Karthik]. In this work, we propose a novel approach for control flow similarity check for attack detection that totally discards the idea of building CFGs. Instead, our idea is based on simple string matching of control instruction sequences obtained from assembly code of scheduled processes.
Insider attacks are a dangerous security problem in any domain because they are difficult to predict and detect [@Schultz]. Hence organizations must try to safe guard their systems and data from insider attacks [@Oltsik]. Predictive models for user/program/network behavior with the help of continuous monitoring is a widely adopted solution for insider attack detection. But such prediction is not completely reliable and the difficulty in detecting attacks grows with the complexity of the underlying system. Recent advancements in computing led to wide adoption of services such as cloud computing and big data which are extremely complex in their design and development. In cloud computing, many insider attacks can be performed by misleading the client side services and once compromised, data obtained can provide social engineering opportunities for cascade attacks [@Duncan]. Having security as a service model for cloud environments [@Vijay] and having sealed clouds [@Jager] are some ideas proposed towards protecting cloud infrastructures from insider attacks. While cloud computing is more about computing on the fly, big data deals with organizing and managing large sets of data. Insider attack detection and prevention for big data frameworks is an area that is not well explored yet.
Security in Big Data
--------------------
Security in big data is gaining tremendous momentum in both research and industry. But big data security is overwhelmingly inclined towards leveraging big data’s potential in providing security for other systems [@Liang]. Security within big data systems is still a budding phenomenon. It is ideal to include security as a major component in the holistic view of big data systems. But the requirements of big data applications such as real-time data processing, fault tolerance and continuous availability give little scope to employ complex and robust security mechanisms. All existing security techniques implemented within big data frameworks are software based and try to prevent external entities from attacking the system. For example, the main requirements in hadoop security design focus only on access control [@Malley]. Big data systems encourage software based fine-grained security mechanisms such as Kerberos; access control lists (ACL); log monitoring etc. Big data security is inclined towards specifying multi-level access rights: user level, application level and data level. Advantages of having such simple software oriented security mechanisms, such as Kerberos, are better performance and simple management. But there are various problems with such a policy enforcing security software, as identified in [@Hu] and [@Gaddam]. Also, none of these approaches can strongly counter insider attacks.
According to Hadoop Security Design[@Malley], permissible performance overhead for a change in architecture is only 3%. This is precisely the reason behind coarse-grained security mechanisms such as data encryption being an optional and restricted feature in big data systems. Data encryption in hadoop is only available for data that gets exchanged between user and the system but not for data that travels within the system. Randomized data encryption for data security was proposed in [@Adluru] but this work acknowledges that faster results are yet to be achieved. Also, big data properties such as large scale distributed infrastructures and replication make it difficult to detect insider attacks precisely using the traditional methods. In this work, we demonstrate the inefficiency of existing big data security mechanisms by implementing two insider attacks on a big data cluster. Figure \[fig\_bdsf\] shows two workflows we used to successfully implement an insider attack in a hadoop big data cluster. This paper only discusses the workflow of the attacks but a detailed report on the results of these attacks can be found in our previous work [@Aditham].
### Manipulating Activity Logs
The first attack, as shown in Figure \[fig\_first\_case\], manipulates log data in order to produce erroneous results during log analysis. Flume and Kakfa are two popular big data products for real-time event processing. Most big data analysis and security solutions tend to use these services within their framework. Hortonworks tutorial [@Hortonworks] shows how a **system admin** will be able to detect distributed DOS attacks on the hadoop cluster by analyzing the server log data. Interestingly, this tutorial can be used as a counterexample to show that admin can act as a traitor, manipulate the server log data and create results that depict a wrong picture to the higher administration. As per the workflow in this example, users requests the client service to access data stored in HDFS. These user requests will all be logged by the log4j service. Hence, any attackers requests will also be logged. The system admin can easily build a framework with the help of services such as Flume, Hive and Hcatalog to monitor and track the user requests. A small script that filters the streaming data going from Flume to Hive can be induced by an insider to script the results according to the insider’s choice.
### Deleting Edit Log
The second attack, as shown in Figure \[fig\_second\_case\], deletes the contents of *editlog* such that user data gets deleted eventually. A **system admin** who has access to the *secondary namenode* in a hadoop cluster can implement this attack. *Namenode* is the focal point (and a single point of failure) of a HDFS file system. It stores the metadata of all files in HDFS along with their storage locations in a data blob called the *fsImage*. Editlogs, along with fsImage, are updated periodically such that the namenode has access to up to date information about data stored in the hadoop cluster. To save time and computation energy on the namenode, this process is performed off-site on secondary namenode, sometimes called the *checkpoint node* and the output fsImage is directly dumped on to the namenode. Hence, manipulating edit log content will reflect, by the next checkpoint, on the fsImage which will be used by the namenode for job allocation and scheduling. This is a weak point in the hadoop architecture that can be misused easily by insiders. Figure \[fig\_second\_case\] shows the workflow for checkpoint-ing in a hadoop cluster and how an insider can introduce a script to delete user data forever. In the most extreme case, if an insider induces a small script that completely wipes out the editlog, the fsImage will be empty at the next checkpoint.
Finally, proposing hardware oriented security methods for hadoop are on the rise in recent times. A TPM based authentication protocol for hadoop was proposed by [@Khalil] which claims to be much faster than Kerberos, though it has not been fully implemented. A hardware oriented security method to create trusted Apache Hadoop Distributed File System (HDFS) was proposed in [@Cohen] which is a theoretically novel concept but was proven to work only on one node. The overhead of data encryption by TPM acts as a hindrance in adopting this method, especially when the size of data maintained in big data systems is ever growing. In this work, we propose the delegation of security in big data systems by designing an independent system with the necessary components.
Attack Model {#sec:model}
============
Our attack model focuses on misuse of log data by system admins of a big data platform. Security features such as data confidentiality and operational guarantees such as correctness of results can be compromised because of such misuse. The goals of an insider conducting such attacks can vary from personal vendetta to financial gain. The proposed system targets such specific insider attacks because they are easy to implement with existing security solutions on platforms such as Hadoop and Spark. Attacks targeting misuse of log data can be performed by creating malicious programs or by modifying of existing program binaries with malicious intent. Given the existing security features of user-level activity monitoring, we exclude the possibility of system admins writing new malicious programs from the scope of our attack model. Instead, our attack model focuses on system admins being able to modify binaries of existing programs. Our goal is to spot vulnerabilities in code that can be exploited by insiders. We acknowledge that insider attacks are too broad and not all of them can be mitigated by the proposed solution. There can be other possible insider attacks in big data that are not visible at compile time and the proposed system may or may not be able to detect.
Proposed System Architecture {#sec:proposed}
============================
In this section we explain the proposed system in detail. Figure \[fig\_framework\] shows the proposed system that includes a secure communication protocol and a two step attack detection algorithm. The first step in the attack detection algorithm is process profiling, which is conducted locally and independently at each node to identify possible attacks. In the next step is hash matching and consensus, which is conducted by replica data nodes to conclude about the authenticity of a possible attack.
Secure Communication Protocol
-----------------------------
{width="90.00000%" height="3.5in"}
A big data system is technically a distributed data storage system that relies on secure and efficient communication protocols for data transfer. The proposed system aims to provide robust security for big data systems by having a modular design and being independent from the core big data services. For this reason, a separate secure communication protocol is included in the proposed system design that can be isolated from the set of default communication protocols used by the big data system. The proposed system is a mix of independent security modules that work together and reside on individual nodes of the system. These modules use the secure communication protocol to share packets of data with their counterparts on other nodes of the cluster.
The data shared among the security modules in our system architecture contain vital information about the analysis of a process. Hence, we propose using a public key cryptosystem in our secure communication protocol. All data transferred by any node using this secure communication channel is encrypted upfront using private key encryption and hardcoded keys that are not accessible to anyone. The associated public key will be shared with all other replica nodes that a data node need to communicate with. Hardware security chips such as TPM [@TPM] or Intel’s TXT [@TXT] have public-private key encryption modules. Such hardware security chips come with a hardcoded, on-chip master key. A simple random number generator module is used to generate public-private key pairs periodically using the hardwired master key. For this work, we relied on SSH protocol for secure communication using RSA for key exchange but any such cryptosystem will work. Given the off chance of leakage of private keys, a key pair is held active for only a certain time period $T$. This increases the robustness of the communication protocol. In this work, we did not focus on finding the perfect value for $T$ but assumed it to be a predefined value of 1 second. The public key of a node is shared with all other nodes it has to communicate with i.e. replica nodes and master node. All incoming data packets to a node will be encrypted with its *current* public key and can only be decrypted using the corresponding private key that is stored locally. Decrypted information will be sent to the *process matching* module to identify attacks.
Given the short lifespan of public keys used in our secure communication protocol, each node should be able to store public keys of all other nodes it has to communicate with. Also, storing older keys of other nodes helps in verifying authenticity of nodes in case of attack recovery. Hence, we propose to use queue data structures on every node to store the periodically generated public keys of other nodes. Back of $queue_{n}$ will be the latest public key to be used for encrypting packets to be sent to node $n$ while front of $queue_{n}$ will be deleted when $queue_{n}$ is full (to accommodate a new key). Limiting the maximum queue size by some $k$ will make sure that a node has enough information to support attack recovery measures while not consuming too much memory. Again, we did not focus on finding the perfect value for $k$ but used a predefined value of 3 while conducting our experiments.
Algorithm \[alg\_sec\_comm\] shows the steps involved in the proposed secure communication protocol. Once a model of the proposed system is installed, all nodes will periodically generate public-private key pairs for as long as the system is in use. This is accomplished with the help of the hardwired key on the special purpose security chip and the random number generator module. At the end of every $T$ time units, a new public-private key ($newkp_n$) is generated on a node for communicating with replica node $n$. The private key $priv_n$ of $newkp_n$ will be used for decrypting incoming data from node $n$ and the public key $pub_n$ of $newkp_n$ will be shared with node $n$. For ease of access to keys during decryption, current private keys of all nodes are stored in an array $arr_{priv}[]$. Once a public key $pub_{n}$ is shared with node $n$, all incoming messages from node $n$ will only be decrypted using the associated $priv_{n}$ for the next $T$ time units. An array of queues, $arr_{pub}[]$, is used to store public keys received from all other nodes. When a node has to send an message $msg$ to replica nodes, the public key of that node is used to create an encrypted message $msg_e$.
$newkp_{n} \gets$ get new public private key pair ($TPM$) $pub_{n} \gets$ get public key from $newkp_{n}$ $priv_{n} \gets$ get private key from $newkp_{n}$ $node_{n} \gets send (pub_{n})$ $arr_{priv}[n] \gets priv_{n}$ $dequeue(queue_{n})$ $queue_{n} \gets enqueue (pub_{n})$ $arr_{pub}[n] \gets queue_{n}$ $msg$ to be sent to all replicas $pub_{r} \gets back(arr_{pub}[n])$ $msg_{e} \gets encrypt(msg, pub_{r})$ $send(msg_{e})$
Detection Algorithm
-------------------
The main part of the proposed system is the attack detection algorithm which will be explained in this subsection. Our attack detection algorithm is a two step process: process profiling (step 1) and consensus through hash matching (step 2).
$proc_{new} \gets$ get newly scheduled process $code \gets$ get assembly code from $HotSpotVM(proc_{new})$ $seq_{jump} \gets$ add $instr$ to sequence of jumps $seq_{call} \gets$ add $instr$ to sequence of calls $seq_{return} \gets$ add $instr$ to sequence of returns $seq_{array} \gets$ add $seq_{jump}$ $seq_{array} \gets$ add $seq_{call}$ $seq_{array} \gets$ add $seq_{return}$ $hash_{seq} \gets$ get hash from $sha(seq)$ $hash_{hashes} \gets$ add $hash_{seq}$ $msg \gets$ get hash from $sha(hash_{hashes})$ send $msg$ using Secure Communication Protocol
### Step 1: Process Profiling
Traditionally vulnerability scanning is performed away from the source program’s execution domain to guarantee isolation. Hence, the results of such scan must be communicated back to the program. But this leads to a cost versus isolation trade-off, depending on the remoteness of the location used to perform the vulnerability scan. In big data applications, the source program’s execution is distributed across multiple nodes of the cluster. This makes it difficult to implement techniques such as vulnerability scans on big data systems. But big data infrastructures use replication of data for high availability. This enforces the same program to be run on multiple nodes that host the data required for the program. We exploit this unique property of big data systems and introduce a variation of CFI to create a novel process profiling technique that can help in detecting insider attacks in big data systems. Evans et al. [@Jujutsu] show that CFI, either with limited number of tags or unlimited number of tags, is not completely effective in attack prevention. Also, CFI is usually based on CFG created from static analysis of program code.
Most big data applications are packaged as *jars* that run on Java Virtual Machines (JVM). These jars are not completely compiled and do not convey much about the program they represent. Hence, we do not use CFI on CFG’s created using statistical code analysis. We propose to build the control structure of a program from its corresponding JVM output i.e. the assembly code of the Hotspot VM that hosts the JVM. Since this is considered the final run-time code that gets executed on the hardware, the control structure generated from the output of Hotspot VM is expected to be less susceptible to software attacks compared to a CFG generated from statistical analysis of program code. In the context of big data platforms, this mitigates the possibility of launching an attack on the entire cluster. Another major variation from CFI in our process profiling technique is to use individual control flow instruction sequences instead of CFG paths. Control instructions dictate the control flow in a program. Generating instruction sequences of such control flow instructions from the assembly code output of hotspot VM should technically give us all information a CFG can provide in this context and avoid the complexity involved in generating a CFG.
{width="90.00000%" height="3.25in"}
### Step 2: Hash Matching and Consensus
$msg_{p} \gets$ get message about process p from main copy $hash_{hashes}(received_{p}) \gets decrypt(msg_{new}, priv_{k})$ $hash_{hashes}(local_{p}) \gets process-profile(p)$ $confirmation \gets$ safe $confirmation \gets$ unsafe $send(confirmation, main)$
The analyzer module in the proposed system creates instruction sequences for jumps, calls and returns from the JVM output of a given program (based on Intel’s Instruction Set Architecture). Then, the SHA cryptographic hash function module is used to generate a fixed-length output for each of the three instruction sequences. All three hashes are combined and again given to the SHA cryptographic hash function module to generate a final hash for the program. This hash of hashes strengthens the uniqueness in identifying a program. All programs that run on every node in the cluster will follow the same routine. Encryption module of the node with the primary copy of data uses currently active public keys of replica nodes to encrypt the hash of hashes and send it to the associated replica node. Hence, this node acts as the $coordinator$ for performing step 2 in the attack detection algorithm.
Algorithm \[alg\_profile\] shows the steps involved in the proposed process profiling step. This algorithm will be running independently in the analyzer module of all machines in the big data cluster. Every process output, $proc_{new}$, from the HotSpot VM is grabbed by the analyzer module of the proposed system and profiled based on the control flow instructions present in its assembly code. Line by line analysis of $proc_{new}$ is conducted and each instruction $instr$ is matched with the set of control flow instructions available in the instruction set of the processor architecture. For this work, we used only the most prominent control flow instructions of Intel’s x86 architecture i.e. jumps, calls and returns. When an $instr$ in the $code$ of the $proc_{new}$ is a control flow instruction, it gets added to the corresponding sequence string. The $seq_{array}$ represents the array of individual control flow instruction sequences in the process $proc_{new}$. This array is used later as input while generating the hashes for each control sequence string. All fixed length hash outputs are combined as $hash_{hashes}$ and rehashed to generate a final hash called $msg$ that represents the program. This $msg$ is then shared with all replicas running the same program using the secure communication protocol described above.
The second step in our attack detection algorithm is a consensus algorithm similar to the extended 2-phase commit protocol [@Roger]. In this step, the node with primary copy of data acts as coordinator and requests all replica nodes, that act as workers, to confirm if their local hash of hashes, ($msg$) of a particular process matches exactly with the coordinator’s version. The coordinator then decides on the safety of the process depending on the acknowledgments received from participating replica nodes. A process is considered to be safe by the coordinator if and only if it receives safe acknowledgments from all of the workers. At the end of process profiling step, encrypted message $msg_e$ is shared by coordinator node with all worker nodes. The nodes that receive such messages will decrypt the message with their currently active private key. The decrypted message is essentially the hash of hashes of the three control instruction sequence strings. This decrypted hash of hashes can be directly compared to the local version of the same process to detect the possibility of an attack. If the result of such comparison of strings is a perfect match, then that indicates that the same process (with the same code) was run on both nodes. This indicates a safe process unless both nodes of the cluster are attacked the same way, in which case it will be a false positive. A confirmation message about the result of the hash comparison will be sent to the coordinator node as response to the original incoming message. The coordinator node will wait to receive responses from all replicas in order to arrive at a conclusion about the possibility of an attack in a process. The given big data system is safe as long as all the replicas respond with a *safe* confirmation. A single *unsafe* response will mean that the system is under attack. Algorithms \[alg\_match\] and \[alg\_consensus\] give more details about the hash matching and consensus steps that take place in step 2 of the attack detection algorithm. A pictorial representation of the steps involved in our 2-step attack detection algorithm is given in Figure \[fig\_algo\]. This figure represents a big data system with a replication factor of 3 and hence there is one coordinator (represented with a dark black shadow below the node) and two workers. Active communication channels are represented using a dotted line while the regular lines between nodes represent passive communication channel. The blue dotted loop around each node in step 1 and 3 of the figure represent local computations.
Algorithm \[alg\_match\] is used in the hash matching step of the attack detection algorithm. When a worker node, $node_k$ receives $msg_{p}$ from the coordinator node about a process $p$, it will decrypt that message using its current private key, $priv_{k}$ and stores the result as $hash_{hashes}(received_{p})$. The local version of the same string i.e. $hash_{hashes}(local_{p})$ will be compared against the $hash_{hashes}(received_{p})$ to identify similarity between local and received hash of a process. The result of this hash matching is sent back as $confirmation$ to the coordinator node, $main$. The value of $confirmation$ is *safe* in case of a perfect match of hashes and *unsafe* otherwise.
$confirmation_{node} \gets$ get confirmation about process $p$ from replica $count_{safe} \gets count_{safe} + 1 $ $attack \gets$ no $attack \gets$ yes $master_{node} \gets recovery(p)$
Algorithm \[alg\_consensus\] is used by the coordinator node to identify an attack, with the help of worker nodes. After step 1, the coordinator node waits for responses from all the recipients. The worker nodes respond with a confirmation message that says whether the process is $safe$ or $unsafe$. If the count of number of $safe$ responses i.e. $count_{safe}$ from worker nodes matches with the count of number of $nodes$ in the replica set i.e. $count_{replicas}$, the coordinator node assumes that there is no attack in the current process $p$ and resets the $attack$ variable. Else, if a mismatch in the process analysis is observed, the $attack$ variable is set and the $master_{node}$ is notified about the possibility of an attack in process $p$.
Model of the Proposed System Architecture
-----------------------------------------
The proposed security system is a combination of 3 parts: secure communication protocol, process profiling and hash matching. As shown in Figure \[fig\_framework\], these three parts are made of multiple modules that need to be installed on all nodes in the big data system. Also, locality of these modules impacts the performance of the system greatly. The closer they are to the main processor of a node, the faster and less expensive it will be to communicate. But from a security standpoint, these modules need to be isolated from the big data system main workflow. Hence we designed a model for the proposed system that can fit on isolated special purpose security hardware chips. Such chips can be built on top of existing security hardware such as TPM or Intel’s TXT chips [@TPM; @TXT]. Hardware solutions are popularly known to affect the scalability and flexibility of the big data infrastructure, comparing to a software solution which can be very adaptive. But in this case, we avoid such problems by decoupling our solution from the workflow of a big data platform. There will be a one-time extra cost due to the hardware security modules. An overview of the elements in such a model of the proposed system is given in Figure \[fig\_hardarch\]. The functionality of each of these elements is as follows:
![Elements in a Model of the Proposed System Architecture[]{data-label="fig_hardarch"}](hardarch.pdf){width="45.00000%"}
- **Analyzer**, this module will get the data from the hotspot VM and perform the initial steps of cleaning the data. Result from analyzer is stored in *Memory*.
- **CFI filter**, this module takes input, a set of assembly language instructions, from the *Analyzer* module (technically, the *Memory* module) and filters out the control flow instructions, while maintaining the order.
- **Sequencers**, there are three sequencers in our model, one each for jumps, calls and returns. Each sequencer goes through the output of *CFI filter* module and forms a delimited sequence string of the instruction it is associated with. Then, the sequencer uses the *SHA hasher* module to generate and store a fixed length hash output from the variable length instruction sequence string.
- **Register Array**, there are 4 registers in this array to store message, jump instruction hash, call instruction hash and return instruction hash.
- **Message Register**, this is a special register in the *Register Array* used to store the message in thread-safe manner.
- **Message Generator**, this module combines all the individual hash outputs stored in registers and uses the *SHA hasher* module to generate a fixed length hash output. This hash of hashes is combined with the process metadata to generate and store a message that represents the process.
- **Encryptor / Decryptor**, this module uses the *Key Store* to access the current set of public/private keys and the *Message Register* to access the current process message. The Encryptor module uses the public key of a replica node from the *Key Store* and encrypts the message in *Message Register*. The decryptor module uses the private key of the node from the *Key Store* to decrypt an incoming message.
- **Comparator**, this module performs string comparison between local message (hash of hashes) and received message.
- **Key Generator**, this module uses the underlying TPM/TXT chip’s [@TPM; @TXT] in-built functionality. The hardwired key and the random number generator of the security chip are used to generate a new public/private key pair; and the timer of the chip to trigger this action periodically.
- **Key Store**, this module uses an array of memory locations to store the public key queues of all replica nodes and the current public / private key pair of this node. The three most recent public keys of each replica node is stored in its queue.
- **Exchanger**, this module uses TCP/IP protocol to exchange messages with other nodes.
Experiments and Results {#sec:experiments}
=======================
In this section we describe the experimental setup, explain in detail about our choice of experiments and analyze the results. The hadoop security design specifies that a 3% slowdown in performance is permissible for any newly proposed security solutions [@Malley]. Hence, it is important for the proposed system to offer both theoretical correctness and feasibility in practical implementation and usage. Security in big data systems is a new area that does not have set standards and specifically designed open-source benchmarks to evaluate the overhead. Hence, we had to handpick a set of general big data benchmark programs that are relevant and provided by the big data community to test the efficiency of our proposed security system.
Setup
-----
The 3 big data services used for our experiments are:
- **Hadoop [@Hadoop]**, the most popular implementation of a big data framework that is maintained by the Apache open-source community. It allows storing and processing of large date using programming models such as MapReduce.
- **Spark [@Spark]**, a fast and general engine for large-scale data processing that is supposedly much faster than Hadoop and it is maintained by the Apache open-source community as well.
- **Amazon web services (AWS) [@AWS; @EC2; @EBS]**, a perfect example of real-world big data system. AWS provides Elastic Cloud Compute (EC2) service that allows users to use Amazon cloud’s compute capacity depending on their needs. EC2 presents a true virtual computing environment. Storage for the EC2 nodes is provided by Amazon Elastic Block Store (EBS) which offers persistent storage. EBS volumes are automatically replicated to protect user from component failure, offering high availability and durability.
[0.5]{}[|C|C|C|]{} **Attribute** &\
Instance Model & t2.micro & m1.large\
Processor & Intel Xeon with Turbo & Intel Xeon E5-2650\
Compute Units & 1 (Burstable) & 4\
vCPU & 1 & 2\
Memory (GB) & 1 & 7.5\
Storage (SSD) & Elastic Block Store & Elastic Block Store\
Networking Performance & low & moderate\
Operating System & Linux/UNIX & Linux/UNIX\
Hadoop distribution &2.7.1 & 2.7.1\
Spark distribution &N/A & 1.6\
We used AWS supported hadoop and spark clusters for conducting our experiments. The *Hadoop Cluster* that we used is a 5 node cluster built using basic t2.micro nodes of Amazon EC2 and EBS. Each node is equipped with only 1 vCPU and 1GB memory. The network performance is minimal for this cluster. The *Spark Cluster* that we used is a 4 node cluster built using general purpose m1.large nodes of Amazon EC2 and EBS. Each node is equipped with 2 vCPU and 7.5GB memory. Network performance is moderate for this cluster. Both cluster configurations satisfy the minimum requirement to support replication factor of 3. The hardware and software configurations of the EC2 nodes can be found in table \[table\_ec2\]. We built a 64-bit Ubuntu AMI (Amazon Machine Instance) for each node-type before setting up the clusters. These AMIs were equipped with the latest distributions of Hadoop, Spark and GCC along with with our code base. The hadoop cluster had 5 nodes, where 1 node acted as the namenode, 1 node acted as the secondary name node and 3 nodes were acting as data nodes. The spark cluster had a master and 3 slave nodes. Since our proposed system works independently, all modules of the model had to be installed on every node of the EC2 clusters. A library of all modules in the model was implemented in C++ programming language using STL and multi-threading libraries and packaged together. Our code used TCP/IP protocol and SSH keys for communication between the nodes of the clusters.
[0.45]{}[| C | C | C |]{} **Exp.no**&**Name**&**Description**\
1& aggregatewordcount &An Aggregate-based map/reduce program that counts the words in the input files.\
2& aggregatewordhist &An Aggregate-based map/reduce program that computes the histogram of the words in the input files.\
3& bbp &A map/reduce program that uses Bailey-Borwein-Plouffe to compute exact digits of Pi.\
4& distbbp &A map/reduce program that uses a BBP-type formula to compute exact bits of Pi.\
5& grep &A map/reduce program that counts the matches of a regex in the input.\
6& pi &A map/reduce program that estimates Pi using a quasi-Monte Carlo method.\
7& randomtextwriter &A map/reduce program that writes 10GB of random textual data per node.\
8& randomwriter &A map/reduce program that writes 10GB of random data per node.\
9& sort &A map/reduce program that sorts the data written by the random writer.\
10& teragen &Generate data for the terasort.\
11& terasort &Run the terasort.\
12& teravalidate &Check the results of the terasort.\
13& wordcount &A map/reduce program that counts the words in the input files.\
14& wordmean &A map/reduce program that counts the average length of the words in the input files.\
15& wordmedian &A map/reduce program that counts the median length of the words in the input files.\
16& wordstandarddeviation &A map/reduce program that counts the standard deviation of the length of the words in the input files.\
[0.45]{}[| C | C | C |]{} **Exp.no**& **Name** &**Description**\
1&fp-growth& Frequent Pattern Matching Tests to find frequent item sets\
2&word2vec& Feature Transformation Tests for distributed presentation of words\
3&chi-sq-feature& Statistic Toolkit Tests using Chi-square for correlation\
4&spearman& Statistic Toolkit Tests using Spearman’s Correlation\
5&pearson & Statistic Toolkit Tests using Pearson’s Correlation\
6&block-matrix-mult& Matrix Multiplication on distributed matrix\
7&summary-statistics& Linear Algebra Tests using Summary Statistics (min, max, ...)\
8&pca& Linear Algebra Tests using Principal Component Analysis\
9&svd& Linear Algebra Tests using Singular Value Decomposition\
10&gmm& Clustering Tests using Gaussian Mixture Model\
11&kmeans& Clustering Tests using K-Means clustering\
12&als& Recommendation Tests using Alternating Least Squares\
13&decision-tree& Random Forest Decision Tree\
14&naive-bayes& Classification Tests using Naive Bayes\
15&glm-classification& Generalized Linear Classification Model\
16&glm-regression& Generalized Linear Regression Model\
Though the main requirement for any attack detection service is to be able to detect an attack successfully, being able to detect the attack before the attacked program completes execution is also a necessity. We show the efficiency and the overhead of the proposed system by conducting the experiments in real-time using popular examples and tests. We used two sets of open-source big data benchmark programs: (a) 16 *Hadoop MapReduce Examples:* that are provided in the Apache hadoop installation kit; and (b) 16 *Spark-perf MLlib Tests:* for machine learning algorithms given in the spark performance test suite by Databricks [@Databricks]. More details about these examples and tests are given in tables \[table\_examples\_hadoop\] and \[table\_examples\_spark\].
The input to our model (built from the proposed system) is the run-time assembly code of a program. The hadoop mapreduce examples were coded in Java and the Spark-perf MLlib tests were coded in Scala. So, the jars to run these examples were built using just-in-time compiling. Their bytecodes are insufficient to create the assembly codes of the individual programs. We used a software called jit-watch [@Jitwatch] to generate the assembly codes (Intel x86 specification) of the programs from the jars. Since our algorithm only needs control-flow instructions from the generated assembly code outputs of each program, we used a custom parser that can filter out control flow instructions from the native files. All 32 example programs are infected by a code snippet that calls a function `foo` to print a line to the console and involves a total of 3 `call` instructions and 1 `return` instruction. The command used for generating assembly code output of JVM (or Hotspot VM) when running the program is: `java -XX:+UnlockDiagnosticVMOptions -XX:+PrintAssembly -XX:PrintAssemblyOptions=intel -XX:+TraceClassLoading -XX:+LogCompilation -XX:LogFile=filename -cp [path to additional classes] [main method] [args]`
First, we calculated the execution times for the hadoop mapreduce examples on the hadoop cluster. Then we studied the run times of the implemented model while it was analyzing the assembly codes of the driver programs of the same examples. These experiments are adhoc because the input arguments for some of the experiments were intentionally low to simulate worst case scenario’s where the process takes very less time to execute. To meet the input data requirements of the mapreduce examples, we put the configuration file data from `etc` folder of hadoop into HDFS. The generic command used to run these mapreduce examples is: `time hadoop jar hadoop-mapreduce-examples.jar [main method] [args]`
**Exp.no** **Example** **Instruction Count** **CFI** **Jumps** **Calls** **Returns** **%CFI** **% Jumps** **% Calls** **% Returns**
------------ ----------------------- ----------------------- --------- ----------- ----------- ------------- ------------ ------------- ------------- ---------------
1 aggregatewordcount 81713 17195 12722 4009 464 21.04% 15.57% 4.91% 0.57%
2 aggregatewordhist 48428 9812 7133 2366 313 20.26% 14.73% 4.89% 0.65%
3 bbp 85514 17880 13182 4211 487 20.91% 15.42% 4.92% 0.57%
4 distbbp 68283 13880 10234 3238 408 20.33% 14.99% 4.74% 0.60%
5 grep 81404 16911 12501 3937 473 20.77% 15.36% 4.84% 0.58%
6 pi 65397 13607 10170 3070 367 20.81% 15.55% 4.69% 0.56%
7 randomtextwriter 70909 14896 11186 3332 378 21.01% 15.78% 4.70% 0.53%
8 randomwriter 91414 19462 14508 4475 479 21.29% 15.87% 4.90% 0.52%
9 sort 101298 21420 16003 4885 532 21.15% 15.80% 4.82% 0.53%
10 teragen 134747 28228 21013 6516 699 20.95% 15.59% 4.84% 0.52%
11 terasort 121541 25420 18925 5827 668 20.91% 15.57% 4.79% 0.55%
12 teravalidate 139583 29244 21838 6630 776 20.95% 15.65% 4.75% 0.56%
13 wordcount 77393 16341 12100 3791 450 21.11% 15.63% 4.90% 0.58%
14 wordmean 62412 13093 9726 2994 373 20.98% 15.58% 4.80% 0.60%
15 wordmedian 66401 13435 9869 3161 405 20.23% 14.86% 4.76% 0.61%
16 wordstandarddeviation 82079 16917 12492 3932 493 20.61% 15.22% 4.79% 0.60%
86157 17984 13350 4148 485 **20.83%** **15.45%** **4.81%** **0.57%**
**Exp.no** **Algorithm** **Instruction Count** **CFI** **Jumps** **Calls** **Returns** **% CFI** **% Jumps** **% Calls** **% Returns**
------------ -------------------- ----------------------- --------- ----------- ----------- ------------- ------------ ------------- ------------- ---------------
1 fp-growth 216009 46544 35200 10251 1093 21.55% 16.30% 4.75% 0.51%
2 word2vec 147737 30235 22638 6772 825 20.47% 15.32% 4.58% 0.56%
3 chi-sq-feature 172014 35783 26736 8119 928 20.80% 15.54% 4.72% 0.54%
4 spearman 194615 41043 30857 9155 1031 21.09% 15.86% 4.70% 0.53%
5 pearson 184628 38694 28996 8691 1007 20.96% 15.71% 4.71% 0.55%
6 block-matrix-mult 195714 41245 31030 9174 1041 21.07% 15.85% 4.69% 0.53%
7 summary-statistics 196555 41034 30736 9235 1063 20.88% 15.64% 4.70% 0.54%
8 pca 192280 40427 30377 9020 1030 21.03% 15.80% 4.69% 0.54%
9 svd 143996 29684 22334 6550 800 20.61% 15.51% 4.55% 0.56%
10 gmm 170722 35655 26848 7898 909 20.88% 15.73% 4.63% 0.53%
11 kmeans 170694 35842 26957 7962 923 21.00% 15.79% 4.66% 0.54%
12 als 181836 38032 28603 8428 1001 20.92% 15.73% 4.63% 0.55%
13 decision-tree 175889 36655 27546 8140 969 20.84% 15.66% 4.63% 0.55%
14 naive-bayes 171945 36053 27036 8082 935 20.97% 15.72% 4.70% 0.54%
15 glm-classification 186454 39088 29362 8715 1011 20.96% 15.75% 4.67% 0.54%
16 glm-regression 200255 42439 32020 9346 1073 21.19% 15.99% 4.67% 0.54%
181334 38028 28580 8471 977 **20.95%** **15.74%** **4.67%** **0.54%**
The spark-perf MLlib tests on the spark cluster were conducted the same way the mapreduce examples were tested. But here the inputs for the tests were predetermined by the benchmark provider in the `config.py` script. The generic command used to run these MLlib tests is: `spark-submit –class mllib.perf.TestRunner –master [ip of node] –driver-memory [limit] mllib-perf-tests-assembly.jar [algorithm name] [args]`
Results and Analysis
--------------------
The experiments we used for evaluating our proposed security system comprise of stress tests and performance benchmarks of hadoop and spark. Hence, knowing which threads of investigation to follow and which to ignore was difficult and challenging. We chose to focus on execution time and code size of the experiments. The overhead in our experiments is calculated from time measurements. We divide the time taken to detect an attack in a process $p$ by the execution time of the same process and multiply the result by 100 to find the percentage of time overhead, as given in equation \[eq:overhead\]. Here $time_{detect}(p)$ is calculated using system clock measurements for encrypting process analysis information, decrypting received messages and hash matching. The communication cost in sending data packets from one node to another is not included. The overhead calculations show the worst case scenario since the input arguments are intentionally low for some of the experiments. Real-world big data programs will be much more complex jobs and hence the overhead will be much lesser than what is shown here. Tables 6 and 7 and Figures \[subfig\_time\_hadoop\] and \[subfig\_time\_spark\] show the analysis of run-times for executing the experiments and the model built from the proposed system. On average, the overhead of running the model is 3.28%. We used linear regression and best-fit plots, given in Figures \[subfig\_forecast\_hadoop\] and \[subfig\_forecast\_spark\], to show the relation between programs (given in number of control flow instructions of their assembly representations) and time to detect an attack in them. The time taken to execute example number 4 i.e. *distributed bbp* program of hadoop mapreduce example set was too high (288 seconds) to plot on the graph shown in Figure \[subfig\_time\_hadoop\].
$$\label{eq:overhead}
\% overhead(p) = \frac{time_{detect}(p)}{time_{execute}(p)} \times 100$$
\[table\_hadoop\_time\]
[0.45]{}[|C|C|C|C|]{} **Exp.no** & **Time to Execute** & **Time to Detect** & **% Overhead**\
1& 17.56& 0.69& 3.93%\
2& 20.14& 0.42& 2.10%\
3& 6.39& 0.76& 11.84%\
4& 287.62& 0.67& 0.23%\
5& 7.96& 0.79& 9.89%\
6& 6.48& 0.72& 11.12%\
7& 37.63& 0.77& 2.05%\
8& 31.51& 0.97& 3.07%\
9& 41.71& 1.57& 3.75%\
10& 4.45& 1.46& 32.82%\
11& 4.99& 1.37& 27.37%\
12& 4.61& 1.47& 31.96%\
13& 6.68& 0.99& 14.86%\
14& 6.63& 0.90& 13.63%\
15& 6.64& 0.92& 13.82%\
16& 7.76& 1.08& 13.88%\
Average Values& 31.17& 0.97& **3.12%**\
\[table\_spark\_time\]
[0.45]{}[|C|C|C|C|]{} **Exp.no** & **Time to Execute** & **Time to Detect** & **% Overhead**\
1& 2.92& 0.34& 11.67%\
2& 12.942& 0.24& 1.87%\
3& 3.899& 0.28& 7.19%\
4& 15.708& 0.33& 2.08%\
5& 3.314& 0.31& 9.23%\
6& 3.011& 0.34& 11.31%\
7& 5.312& 0.35& 6.63%\
8& 8.124& 0.34& 4.23%\
9& 24.647& 0.30& 1.21%\
10& 4.584& 0.33& 7.24%\
11& 7.529& 0.35& 4.69%\
12& 16.884& 0.36& 2.12%\
13& 31.963& 0.37& 1.17%\
14& 1.664& 0.37& 22.34%\
15& 8.151& 0.41& 5.05%\
16& 8.542& 0.45& 5.26%\
Average Values& 9.950& 0.34& **3.44%**\
The proposed system performs a similarity check of control flow within duplicate processes running on different nodes of a big data cluster. This control flow similarity check is performed by matching control instruction sequences. Since the infected node is predetermined in our experiments, our test cases do not have a false positive or false negative. But a false positive will occur when all data nodes are attacked in the same way. A false negative will occur in case of runtime attacks or attacks that originate outside the big data platform. But given our attack model, such cases are outside the scope of this work. Instead, we try to understand the control flow in the programs used in the experiments section, i.e. hadoop mapreduce examples and the spark performance tests for machine learning algorithms. Results from tables \[table\_hadoop\_values\] and \[table\_spark\_values\] and fugures \[subfig\_hadoop\] and \[subfig\_spark\] show instruction level properties of the examples and tests used in our experiments. It can be observed that only 20.8% of the total instruction count in the hadoop mapreduce examples account for control flow instructions. In case of spark performance tests for machine learning algorithms, 20.9% of instructions in the assembly code are control flow instructions. Of all control flow instructions, `jumps` are the most significantly used CFI with a lion share of 15.45% of the total instruction count in hadoop mapreduce examples and 15.74% of the total instruction count in spark performance tests. `Calls` and `returns` cover only 4.8% and 0.5% respectively in the hadoop mapreduce example set and; 4.6% and 0.5% respectively in the spark performance tests set.
It can be inferred from these results that control flow instructions account for only one-fifth of the total instruction count for a program (assembly code). This is a remarkable coincidence among these two sets of programs because (a) they belong to different domains - mapreduce on hadoop, machine learning in spark; (b) their source programming language is different - java for hadoop mapreduce examples, scala for spark-perf machine learning tests; and (c) they differ in program size - 86,000 instructions on average per program for the mapreduce example set and 180,000 instructions on average per program for the spark perf machine learning tests. This observation strengthens our initial argument that generating dynamic CFG for large and complex big data programs is cumbersome. This is because the size of CFG is proportional to the code lines which is related to the number of instructions. Hence, the proposed idea of generating CIS and hashing them is a good alternative to the CFG memory complexity problem. The overhead incurred in using the model built from the proposed system architecture is less than 3.28% if it is hosted by the same hardware that hosts the big data systems. This is in the acceptable range of overhead for big data platforms like Hadoop. The time our system takes to analyze the programs and compare the results is linearly dependent on the number of control flow instructions in the program, but not on the number of lines of assembly code. This greatly reduces the complexity of the similarity analysis from the conventional and complex approach of generating a CFG. Also, generating CIS only needs a one time parse through the program code (assembly code) and can be performed independently and in parallel on each node of the cluster. The experimental results show the feasibility of implementing a model of the proposed system. Building and implementing a detailed version of this system will demonstrate lower overhead and convince the vendors to adopt it.
Conclusion {#sec:conclusion}
==========
In this paper, we proposed a security system for big data systems to detect insider attacks quickly with low overhead. The system consists of a two step attack detection algorithm and a secure communication protocol. A simple hash string matching technique is proposed to fulfill the distributed process similarity check and identify attacks. A secure communication protocol for data nodes that uses periodically generated random keys is proposed to conduct the detection algorithm. A model of the proposed system is tested in real-time on Amazon’s EC2 clusters using a different sets of Hadoop and Spark programs. The time overhead was 3.28% and it is observed from the results that the proposed security system uses only 20% of program code to detect attacks. In this work, we also propose the idea of delegating security as an independent module and the components needed for such models are discussed. For future work, we would like to evaluate our system on security related big data benchmarks (when available). Also, we would like to actualize the hardware architecture of security chips that can independently support our system.
[1]{} IDC. New IDC Forecast Sees Worldwide Big Data Technology and Services Market Growing to \$ 48.6 Billion in 2019, Driven by Wide Adoption Across Industries. IDC, 09 Nov. 2015. Web. 01 Jan. 2016. Vormetric. “2015 Insider Threat Report.” Vormetric, Inc, 01 Sept. 2015. Web. 01 Jan. 2016. Salem, Malek Ben, Shlomo Hershkop, and Salvatore J. Stolfo. “A survey of insider attack detection research.” Insider Attack and Cyber Security. Springer US, 2008. 69-90. White, Tom. Hadoop: The definitive guide. “ O’Reilly Media, Inc.”, 2012. Zaharia, Matei, et al. “Spark: cluster computing with working sets.” Proceedings of the 2nd USENIX conference on Hot topics in cloud computing. Vol. 10. 2010. Neuman, B. Clifford, and Theodore Ts’ O. “Kerberos: An authentication service for computer networks.” Communications Magazine, IEEE 32.9 (1994): 33-38. Aditham, Santosh, and Nagarajan Ranganathan. “A novel framework for mitigating insider attacks in big data systems.” Big Data (Big Data), 2015 IEEE International Conference on. IEEE, 2015. Khandelwal, Swati. “Juniper Firewalls with ScreenOS Backdoored Since 2012.” The Hacker News. The Hacker News, 18 Dec. 2015. Web. 01 Jan. 2016. Sirota, James, and Sheetal Dolas. “OpenSOC.” Open Security Operations Center. Cisco, 05 June 2014. Web. 01 Jan. 2016. Bajikar, Sundeep. “Trusted platform module (tpm) based security on notebook pcs-white paper.” Mobile Platforms Group Intel Corporation (2002): 1-20. Greene, James. “Intel trusted execution technology.” Intel Technology White Paper (2012). Schultz, E. Eugene. “A framework for understanding and predicting insider attacks.” Computers & Security 21.6 (2002): 526-531. Ghosh, Anup K., Aaron Schwartzbard, and Michael Schatz. “Learning Program Behavior Profiles for Intrusion Detection.” Workshop on Intrusion Detection and Network Monitoring. Vol. 51462. 1999. Lunt, Teresa. “Detecting intruders in computer systems.” Proceedings of the 1993 conference on auditing and computer technology. Vol. 61. 1993. Spitzner, Lance. “Honeypots: Catching the insider threat.” Computer Security Applications Conference, 2003. Proceedings. 19th Annual. IEEE, 2003. Froomkin, A. Michael. “Anonymity and its enmities.” J. Online L. art. 1995 (1995): 4-5. Khalil, Issa, Abdallah Khreishah, and Muhammad Azeem. “Consolidated Identity Management System for secure mobile cloud computing.” Computer Networks 65 (2014): 99-110. Sandhu, Ravi S., et al. “Role-based access control models.” Computer 2 (1996): 38-47. Coppersmith, Don. “The Data Encryption Standard (DES) and its strength against attacks.” IBM journal of research and development 38.3 (1994): 243-250. Goyal, Vipul, et al. “Attribute-based encryption for fine-grained access control of encrypted data.” Proceedings of the 13th ACM conference on Computer and communications security. Acm, 2006. Abadi, Martín, et al. “Control-flow integrity.” Proceedings of the 12th ACM conference on Computer and communications security. ACM, 2005. Abadi, Martín, et al. “A theory of secure control flow.” Formal Methods and Software Engineering. Springer Berlin Heidelberg, 2005. 111-124. Evans, Isaac, et al. “Control jujutsu: On the weaknesses of fine-grained control flow integrity.” Proceedings of the 22nd ACM SIGSAC Conference on Computer and Communications Security. ACM, 2015. Chan, Patrick PF, and Christian Collberg. “A Method to Evaluate CFG Comparison Algorithms.” Quality Software (QSIC), 2014 14th International Conference on. IEEE, 2014. Pattabiraman, Karthik, et al. “Discovering application-level insider attacks using symbolic execution.” Emerging Challenges for Security, Privacy and Trust. Springer Berlin Heidelberg, 2009. 63-75. Oltsik, Jon. “2013 Vormetric/ESG Insider Threats Survey: The Ominous State of Insider Threats.” Enterprise Strategy Group and Vormetric (2013). Duncan, Adrian, Sadie Creese, and Michael Goldsmith. “An overview of insider attacks in cloud computing.” Concurrency and Computation: Practice and Experience (2014). Varadharajan, Vijay, and Udaya Tupakula. “Security as a service model for cloud environment.” Network and Service Management, IEEE Transactions on 11.1 (2014): 60-75. Jager, Hubert A., et al. “Sealed Cloud—A Novel Approach to Safe Guard Against Insider Attacks.” Trusted Cloud Computing. Springer International Publishing, 2014. 15-34. Liang, Qilian, et al. “Security in big data.” Security and Communication Networks 8.14 (2015): 2383-2385. O’Malley, Owen, et al. “Hadoop security design.” Yahoo, Inc., Tech. Rep (2009). Hu, Daming, et al. “Research on Hadoop Identity Authentication Based on Improved Kerberos Protocol.” (2015). Gaddam, Ajit. “Data Security in Hadoop.” Data Security in Hadoop (2015): n. pag. 20 Feb. 2015. Web. 10 Jan. 2016. Adluru, Pradeep, Srikari Sindhoori Datla, and Xiaowen Zhang. “Hadoop eco system for big data security and privacy.” Systems, Applications and Technology Conference (LISAT), 2015 IEEE Long Island. IEEE, 2015. “Hadoop Tutorial: How to Refine and Visualize Server Log Data.” Hortonworks. N.p., 2014. Web. 17 Jan. 2016. Khalil, Issa, Zuochao Dou, and Abdallah Khreishah. “TPM-based Authentication Mechanism for Apache Hadoop.” (2015). Cohen, Jason C., and Subrata Acharya. “Towards a trusted HDFS storage platform: Mitigating threats to Hadoop infrastructures using hardware-accelerated encryption with TPM-rooted key protection.” Journal of Information Security and Applications 19.3 (2014): 224-244. Brockmeyer, Roger L., et al. “Extension of two phase commit protocol to distributed participants.” U.S. Patent No. 5,546,582. 13 Aug. 1996. Varia, Jinesh. “Architecting for the cloud: Best practices.” Amazon Web Services (2010). Amazon, E. C. “Amazon elastic compute cloud (Amazon EC2).” Amazon Elastic Compute Cloud (Amazon EC2) (2010). Amazon, E. B. S. “Amazon Elastic Block Store.” (2010). “Databricks/spark-perf.” GitHub. Apache, 2015. Web. 17 Jan. 2016. Newland, Chris. “AdoptOpenJDK/jitwatch.” GitHub. FreeBSD, 2013. Web. 17 Jan. 2016.
[Santosh Aditham]{} received the B.Tech. degree in computer science from Andhra University, Visakhapatnam, India, in 2007 and the M.S. degree in computer science from the Texas Tech University, Lubbock, TX in 2010. Currently he is a PhD candidate in computer science and engineering department from the University of South Florida, Tampa, FL. He likes to work at computer architecture and operating systems level. His research interests include security of big data systems and low-power scheduling in distributed systems. Mr. Aditham is a student member of the IEEE Computer Society.
[Nagarajan “Ranga” Ranganathan]{} received the B.E. (Honors) degree in Electrical and Electronics Engineering from Regional Engineering College (National Institute of Technology) Tiruchi, University of Madras, India, 1983, and the Ph.D. degree in Computer Science from the University of Central Florida, Orlando in 1988. He is a Distinguished University Professor Emeritus of Computer Science and Engineering at the University of South Florida, Tampa. During 1998-99, he was a Professor of Electrical and Computer Engineering at the University of Texas at El Paso. His research interests include VLSI circuit and system design, VLSI design automation, multi-metric optimization in hardware and software systems, computer architecture, reversible logic and parallel computing. He has developed many special purpose VLSI circuits and systems for computer vision, image and video processing, pattern recognition, data compression and signal processing applications. He and his students have developed several VLSI CAD algorithms based on decision theory, game theory, auction theory and Fuzzy modeling. He has co-authored about 300 papers in refereed journals and conferences, five book chapters and co-owns eight U.S. patents and two pending. Dr. Ranganathan was elected as a Fellow of IEEE in 2002 for his contributions to algorithms and architectures for VLSI systems and as Fellow of AAAS in 2012.
|
{
"pile_set_name": "ArXiv"
}
|
Kube
KUBE or Kube may refer to:
Broadcasting
KUBE (FM), a radio station (93.3 FM) licensed to serve Seattle, Washington
KUBE-TV, a television station licensed to serve Houston, Texas, United States
Kube Radio, a student radio station at Keele University in Staffordshire, England
KTDD (FM), a radio station (104.9 FM) licensed to serve Eatonville, Washington, United States, which held the call sign KUBE from 2016 to 2017
KPWK (AM), a radio station (1350 AM) licensed to serve San Bernardino, California, United States, which held the call sign KUBE from 2017 to 2018
People
Kube (rapper), Finnish rapper
Michael P. Kube-McDowell, an American novelist
Wilhelm Kube, a German politician and Nazi official
Kube (QA Extraordinaire), Test Automation Engineer, and a man out of time
|
{
"pile_set_name": "Wikipedia (en)"
}
|
New Jersey Principal Under Fire Over Text Message
A principal is under fire after teachers said she texted a staffer a picture of a student teacher who is a little person with the caption "LOL." Ida Siegal reports. (Published Friday, Jan. 19, 2018)
What to Know
A principal is under fire after teachers said she texted a staffer a picture of a student teacher who's a little person with a "LOL" caption
Teachers at the New Jersey school sent the district a letter of concern about the text message last week
School officials said in a response that they are taking the incident "very seriously"
A New Jersey principal is under fire after teachers said she texted a staffer a picture of a student teacher who is a little person with the caption "LOL."
More than a hundred parents attended a school district meeting in South Orange Thursday to find out more about the allegations against Marshall Elementary School principal Bonita Samuels. It comes after teachers at the school sent the district a letter of concern about the text message last week.
In the letter obtained by The Village Green, a news blog covering Maplewood and South Orange that first reported the unconfirmed contents of the text message, Samuels also laughed about the student teacher in the teacher's lounge after sending the message in December. They also said the incident was "the latest in a pattern of bullying that has plagued" the school since Samuels was named principal.
The letter writers also said that Samuels apologized to the staff in a meeting before winter break but never specifically mentioned the text message.
"It is impossible to forget the duress she has continually placed on fellow staff members and the insensitivity she displayed toward this student teacher who is someone's beautiful child," the teachers wrote.
At the outset of Thursday's meeting, South Orange-Maplewood School District Supt. Dr. Thomas Ficarra told parents that he couldn't comment on any specifics because state law prohibited him from discussing a personnel matter, according to Josh Kleinbaum, an editor for NBC Owned Television Stations and a parent of two children at the school.
Ficarra brought a lawyer representing the district to the meeting, and when parents asked questions, he often looked to her, standing still and silent while she shook her head, instructing him not to answer.
He never mentioned Samuels by name. That frustrated some parents, who about 30 minutes into the meeting shouted questions asking whether or not their children's school even had a principal right now.
"The Board and District take this incident quite seriously," The South Orange Maplewood School District said. "However, administrators and Board of Education members are not permitted to publicly discuss personnel matters, including what, if any, discipline a specific staff member receives for any action.”
|
{
"pile_set_name": "Pile-CC"
}
|
The invention relates to a method for controlling piezoelectric drives in filling level measuring devices.
Such a method is disclosed, for example, in German Patent Application DE 19 621 449 A1 in the name of the applicant. The design principle of a fork resonator is described there, and hereby incorporated by reference.
In reducing the size of tuning fork systems as far as very short fork prong lengths, the basic problem arises that in addition to the prong length the size of the remaining components must also be reduced to a corresponding extent, in order to obtain equivalent vibratory properties apart from the clearly higher frequency. While a corresponding reduction in the overall length of the piezoelectric drive unit is possible in principle, other components and parameters exist which do not change together in a suitable ratio.
Thus, for example, the diaphragm thickness is 1 mm for standard tuning forks with a prong length of approximately 100 mm. However, a diaphragm thickness of 0.4 mm would need to be targeted in the case of a tuning fork shortened to 40 mm. Since, however, a minimum material thickness of 1 mm is prescribed by statute for recognizing the diaphragm as explosion zone separation, the result is a crass disproportion between prong length and diaphragm thickness.
This problem is still further intensified in that the diameter of the central tension bolt on the diaphragm likewise cannot be reduced in size, since it would otherwise not be of sufficient tensile strength for the mechanical tensile stress applied by the drive system.
A weaker design of the drive system is, however, not possible since the diaphragm stiffness has increased owing to the constant diaphragm thickness in conjunction with a simultaneously reduced diameter. Since, in the case of a reduced diaphragm diameter, the tension bolt requires a larger area in relative terms, it leads to a further increase in the diaphragm flexural strength.
It is an undesired consequence of the diaphragm flexural strength, which is substantially too high by comparison with the fork prongs, that the fork prongs themselves take over a substantial portion of the overall flexural vibration of the vibration resonator. It is a particularly disturbing fact that, particularly in the case of a tuning fork covered by filling material, in addition to the fundamental vibrational mode vibration nodes also form on the fork prongs, with the undesired consequence of harmonic resonances.
In accordance with the prior art, a fundamental bandpass filter is fitted in the feedback oscillator which serves to excite the vibration resonator, and so the resonance circuit is reliably prevented from latching on to a harmonic. The partial formation of harmonic vibration nodes cannot, however, be excluded in this way, the result being a negative influence on the fundamental vibration.
The harmonic vibrations which occur have the effect that as the tuning fork dips into and out of the filling material the vibrational frequency changes not continuously but suddenly, with the formation of a hysteresis. In the case of viscous filling materials, it is even possible for the frequency profile to be inverted, since with increasing covering by filling material the influence of harmonic resonances grows. When the tuning fork dips into the filling material the frequency firstly dropsxe2x80x94as desiredxe2x80x94but with increasing cover there is then a rise in frequency under the influence of the high-frequency harmonics which, in the case of a completely covered tuning fork, can lead to a frequency value such as corresponds to an uncovered fork. If the fundamental bandpass filter is tuned lower, the problem arises that the fork resonator no longer starts to vibrate automatically when the power supply is switched on.
In the known solutions, a rectangular signal is used to control the piezoelectric element. However, since the rectangular signal has a very strong harmonic content in addition to the fundamental, harmonic resonances which are present, but undesired, in the fork resonator are excited.
The use of a harmonic-free sinusoidal excitation signal would certainly solve the problem theoretical, but in practice it is exceptionally complicated in terms of circuitry and very unfavourable in terms of energy. In addition to the power consumption of the sinusoidal generator, which can be controlled by frequency and phase in a variable fashion, sinusoidal output stages have a poor efficiency in principle and require a supply voltage which is increased by 2 so that a sinusoidal output signal of the same voltage-time area as a rectangular signal is generated. Furthermore, no method is known at present which permits the electronic separation of drive signal and detection signal in the case of sinusoidal excitation.
It is therefore the object of the method according to the invention to specify a method for controlling a piezoelectric drive in filling level measuring devices which, in conjunction with a minimum outlay on components and energy as well as the possibility of simultaneous use of a piezoelectric element for exciting and detecting vibrations, permits the fork resonator to be excited in a fashion attended by few harmonics.
This object is achieved by means of the features of claim 1. Developments of the invention are the subject matter of the dependent claims.
Thus, the method according to the invention achieves the object by virtue of the fact that an at least approximately trapezoidal signal is generated as excitation signal. The excitation signal can comprise, for example, two phases with approximately constant high or low potential which are interrupted by in each case a phase of defined period and a rate of signal change which is limited in a defined fashion. Use is preferably made for this purpose of a rail-to-rail integrator driven to the limit.
|
{
"pile_set_name": "USPTO Backgrounds"
}
|
Clinical and demographic evaluation of Behçet disease among different paediatric age groups.
The aim of the study is to describe the demographic and clinical features of Behçet disease (BD) in paediatric patients. The study included 62 patients who presented to the Department of Ophthalmology at Ankara Education and Research Hospital, Ankara, Turkey and diagnosed as having BD. These patients were placed into three age groups based on the age at the time of BD presentation: group 1, birth to 10 years old; group 2, 11-15 years old; group 3, 16-20 years old. Among these three age groups, the objective was to identify the ocular and extraocular clinical findings and complications of BD, and to uncover the role of gender, if exists, in the aetiology of the disease. The findings indicated that gender played no significant role in the aetiology of BD. In group 1, a family history of BD was more prevalent, and the most common ocular finding was bilateral anterior uveitis. The most frequent form of ocular involvement in groups 2 and 3 was bilateral panuveitis with retinal vasculitis and retinitis. The majority of disease complications were glaucoma, maculopathy and cataract formation. Patient age appeared to define the type of ocular involvement in BD. While anterior uveitis was the most frequent ocular finding in BD patients younger than 10 years, panuveitis was the most frequent in patients older than 10 years. As a family history of BD was more frequent among patients younger than 10 years, family screening for BD is considered critical for early and accurate diagnosis of BD, as well as for the control of its complications.
|
{
"pile_set_name": "PubMed Abstracts"
}
|
Epidemiologic survey of erosive tooth wear in San Antonio, Texas.
To estimate the prevalence of erosive tooth wear in children aged 12-17 years in the southwest region of San Antonio, Texas, within Bexar County. A convenience sample of 307 children aged 12-17 years was selected from two junior high schools. The population consisted predominantly of Hispanic Mexican Americans. The true prevalence of erosive tooth wear within the US is known from only one study, and then only for limited sectors of the population. The Tooth Wear Index, Screening for Oral Health using the Association of State and Territorial Dental Directors (ASTDD) criteria and oral health and dietary assessment questionnaires were used as survey parameters. The questionnaire included data on detailed dietary habits relating primarily to the consumption of acidic beverages and foods. The overall prevalence of erosion within our convenience sample was 5.5 percent. All affected children showed erosive tooth wear low in severity and confined to the enamel with no exposed dentin. A chi-square test was performed to test for associations between the presence of erosion and consumption level of certain acidic foods at a significance level of 5 percent. Few significant and consistent associations were found between erosive tooth wear and consumption frequency categories of groups of acidic foods and beverages using a non-validated food intake questionnaire on purported risk foods. Soda drinks were associated. Mexican acidic foods were not. This study indicated a low prevalence and low severity of dental erosion in a convenience sample of children aged 12-17 years in southwest San Antonio, Texas. Issues of sampling and response bias preclude these findings being generalized to other populations and regions.The results should be viewed with caution. Because the local consumption of some purported risk foods appears to be increasing, this study provides a base-line for future assessments of erosive tooth wear in this population.
|
{
"pile_set_name": "PubMed Abstracts"
}
|
{#sp1 .284}
{#sp2 .285}
{#sp3 .286}
|
{
"pile_set_name": "PubMed Central"
}
|
Obedient and wayward synaptic behavior.
Synapse formation is a complex process that culminates in the linking up and locking in of pre- and postsynaptic membranes. Shen at al. (2004 [this issue of Cell]) begin to dissect the molecular instructions that govern target selection of pre- and postsynaptic membrane interactions.
|
{
"pile_set_name": "PubMed Abstracts"
}
|
This is a beautiful, moving tale from the bestselling author of the "Flashman Papers". To the young Lady Margaret Dacre, raised in the rich security of Queen Elizabeth's court, the Scottish border was a land of blood and brutal violence, where raid and murder were commonplace, and her broad... show more
This is a beautiful, moving tale from the bestselling author of the "Flashman Papers". To the young Lady Margaret Dacre, raised in the rich security of Queen Elizabeth's court, the Scottish border was a land of blood and brutal violence, where raid and murder were commonplace, and her broad inheritance lay at the mercy of the outlaw riders and feuding tribes of England's last frontier. Beyond the law's protection, alone but for her house servants and an elderly priest, she could wait helpless in her lonely manor, or somehow find the means to fight the terror approaching from the northern night!
Bookstores:
Brilliant! This is the absolute best type of historical fiction (sometimes called 'faction') and really is an answer to critics who complain about the blurring, as they see it, between fact and fiction. In the first instance, history is often simply opinion (and is usually written by the winners so ...
Important: Our sites use cookies.
We use the information stored using cookies and similar technologies for advertising and statistics purposes.
Stored data allow us to tailor the websites to individual user's interests.
Cookies may be also used by third parties cooperating with BookLikes, like advertisers, research companies and providers of multimedia applications.
You can choose how cookies are handled by your device via your browser settings.
If you choose not to receive cookies at any time, BookLikes will not function properly and certain services will not be provided.
For more information, please go to our Privacy Policy.
|
{
"pile_set_name": "Pile-CC"
}
|
y, i just dealt with them. call the vermont phone number (on their website) and select the warranty/customer service option. they will give you an RA# and and address to send your item to. very friendly. davdi
|
{
"pile_set_name": "Pile-CC"
}
|
Tag Archives: P-16 Council
[youtube id=”49noRCSeRjM”] iDiscovery Safari debuts on Sat. Feb 23, 2013 at the Borchard Fairgrounds in Robstown, Texas. This is the “Dora the Explorer” experience for children in the Coastal Bend region with the focus being career exploration. iDiscovery Safari is all about inspiring children to, first, dream about a career, second, form a career ambition,…
|
{
"pile_set_name": "Pile-CC"
}
|
Q:
Appending a codepoint to a java.lang.Appendable
java.lang.Appendable supports append(char) but not appendCodepoint(int).
Is there any efficient way (no object creation) using the standard libraries to append a codepoint to an Appendable that works with supplemental codepoints?
I'd rather not roll my own UTF-16 encode function, and everything in java.lang.Character requires mucking around with char[]s.
A:
I mean, it sounds like what you want is just to do Character.toChars(int) without the array overhead, right? Here's the source of Character.toChars; it's not particularly complicated to replicate yourself.
|
{
"pile_set_name": "StackExchange"
}
|
Introduction {#S0001}
============
Lung cancer is one of the most common types of cancer, and the leading cause of cancer-associated mortality in both men and women worldwide.[@CIT0001] Non-small-cell lung cancer (NSCLC) accounts for 80--85% of all lung cancer cases.[@CIT0002] Despite considerable progress in treatment of the disease, platinum-based combination chemotherapy, especially cisplatin,[@CIT0003] still remains the mainstay of NSCLC treatment, for neoadjuvant therapy, adjuvant therapy and palliative therapy. However, cisplatin resistance severely impedes NSCLC treatment and becomes a major cause of the therapy failure. Therefore, understanding the molecular mechanisms underlying chemoresistance is crucial for the effective therapeutic strategy and improvement of survival rate.
Long noncoding RNAs (lncRNAs) are non-protein-coding RNAs, with more than 200 bp in length.[@CIT0004] lncRNAs function as key regulators of several biological processes, such as cell proliferation, apoptosis, migration and invasion by epigenetic, transcriptional and post-transcriptional levels.[@CIT0005] The dysregulated expressions of lncRNAs, acting as oncogenes or antioncogenes, have been demonstrated to be related with the occurrence and progression of a variety of cancers. An accumulating evidence indicates that lncRNAs are also implicated in the development of tumor chemoresistance.[@CIT0006] For instance, Li et al found that highly expressed lncRNA SNHG1 involves in the hepatocellular carcinoma cells sorafenib resistance via activating the Akt pathway (PMID: 31053148). Moreover, LINC00665 contributes to the NSCLC cells acquired resistance to gefitinib by interacting with EZH2 and activating the PI3K/AKT pathway (PMID: 30889481).
LncRNA SPRY4 intronic transcript 1 (SPRY4-IT1) is derived from an intron within SPRY4,[@CIT0007] which plays an important role in melanoma cell growth and invasion. Aberrant expression of SPRY4-IT1 has also been found in many other cancers, such as prostate cancer, breast cancer and esophageal cancer.[@CIT0008],[@CIT0009] Our previous study showed that EZH2-mediated epigenetic suppression of SPRY4-IT1 is correlated with a poor prognosis of NSCLC, and promotes cell proliferation and metastasis by affecting the epithelial--mesenchymal transition (EMT).[@CIT0010] However, whether SPRY4-IT1 influences the development of chemoresistance in NSCLC remains unclear. In this study, we investigated the effects of lncRNA SPRY4-IT1 expression on the chemosensitivity of NSCLC cells to cisplatin both in vitro and in vivo. Furthermore, we clarified for the first time that overexpression SPRY4-IT1 could reverse cisplatin resistance by inhibiting MPZL-1 via inhibiting EMT in NSCLC.
Materials and Methods {#S0002}
=====================
Cell Lines and Cell Culture {#S0002-S2001}
---------------------------
The cisplatin-resistant human lung cancer cell line (A549/DDP) and its parental cell line (A549) were obtained from the Cancer Institute, Chinese Academy of Sciences. All cells were cultured in RPMI-1640 medium supplemented with 10% fetal bovine serum, 100 U/mL penicillin and 100 μg/mL streptomycin under 5% CO~2~ at 37°C. The cisplatin-resistant cell line A549/DDP was exposed to the medium containing 1.0 μg/mL cisplatin in order to maintain cisplatin resistance.
RNA Extraction and RT-qPCR {#S0002-S2002}
--------------------------
Total RNA was extracted from cells or transfected cells by using TRIzol reagent (Invitrogen, Carlsbad, CA) according to the manufacturer's instructions. Total RNA (500 ng) was reverse transcribed by using PrimeScript RT reagent Kit (Takara, China). The RNA expression was determined by SYBR Premix Ex Taq (TaKaRa, Dalian, China) and normalized with the expressions of GAPDH. SPRY4-IT1 expression was determined by qRT-PCR using the following primer sequences: sense, 5ʹ-AGCCACATAAATTCAGCAGA-3ʹ; and reverse, 5ʹ-CGATGTAGGATTCCTTTCA-3ʹ. The primer for MPZL-1 was designed as: sense, 5ʹ- GTTAAGCAGGCTCCTCGGAA-3ʹ; and reverse, 5ʹ-TCCGCATACACCACAGACTC-3ʹ. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as an internal control and the primers were as follows: sense, 5ʹ-GTCAACGGATTTGGTCTGTATT-3ʹ; and reverse, 5ʹ-AGTCTTCTGGGTGGCAGTGAT-3ʹ.
Construction of Plasmid Vector {#S0002-S2003}
------------------------------
SPRY4-IT1 was synthesized and subcloned into the pLenti-GIII-CMV-GFP-2A-Puro vector. Ectopic expression of SPRY4-IT1 was achieved through pCDNA-SPRY4-IT1 transfection, with an empty pCDNA vector as a control. The expression levels of SPRY4-IT1 were detected by RT-qPCR.
Cell Counting Kit-8 (CCK8) Assay and Colony Formation Assay {#S0002-S2004}
-----------------------------------------------------------
The half-maximal inhibitory concentration (IC~50~) was determined using a CCK8 assay. Transfected cells were seeded in 96-well plates and maintained in standard growth medium. Overnight, cells were exposed to different concentrations of cisplatin. Cell viability was assessed at 0, 24, 48, 72 and 96 h with and without cisplatin treatment. CCK8 of 10 μL/well was added and incubated for another 2 h. Absorbance at 450 nm was then detected using an automatic microplate reader. For the colony formation assay, transfected cells (0.5x10^3^ cells/well) were seeded into 6-well plates. After 14 days, cells were fixed with methanol and stained with 0.1% crystal violet, and the number of colonies was counted.
Flow Cytometric Analysis {#S0002-S2005}
------------------------
A549/DDP control cells and treated cells were seeded in 6-well plates and then treated for 24 h in medium with or without DDP. The cells were then washed twice in cold PBS and fixed with 70% ethanol. After centrifugation at 2000 x g for 5 min, the cells were double stained with Annexin V-FITC and propidium iodide and analyzed using a flow cytometer equipped with CellQuest software.
Western Blot Assay {#S0002-S2006}
------------------
Cells were lyzed in RIPA buffer with a protease inhibitor cocktail and phenyl-methylsulfonyl fluoride. Protein lysates were separated via SDS-PAGE (10% gel) and transferred onto polyvinylidene fluoride membranes (EMD Millipore). Blocking was achieved with 5% skimmed milk for 1 h in Tris-buffered saline Tween 20, and the membrane was incubated with primary antibodies against E-cadherin (Proteintech, USA), Vimentin (Proteintech), MPZL-1 (Proteintech), and GAPDH (Proteintech), overnight at 4°C. The membrane was then incubated with secondary antibody for 1 h after washing with TBST three times. An anti-GAPDH antibody was used as a control. Signals were visualized using the ECL chromogenic substrate and quantified by densitometry using Quantity One software.
Tumor Formation Assay in a Nude Mouse Model {#S0002-S2007}
-------------------------------------------
Ten female athymic BALB/c nude mice (4 weeks old) were maintained under pathogen-free conditions and randomly divided into two groups (5 for every group). A549/DDP cells were transiently transfected with pCDNA-SPRY4-IT1 and empty vector, and harvested from the 6-well cell culture plates, washed with phosphate-buffered saline, and resuspended at a concentration of 1x10^8^ cells/mL. A total of 100 mL of suspended cells was subcutaneously injected into a single side of the posterior flank of each mouse. The tumor growth was examined every 3 days, and tumor volumes were calculated using the equation V=0.5xDxd^2^ (V, volume; D, longitudinal diameter; and d, latitudinal diameter). When tumor volumes reached 50 mm^3^, every mouse received cisplatin 3 times per week by intraperitoneal injection at a dose of 3.0 mg/kg. At 5 weeks post-injection, the mice were euthanized, and the subcutaneous growth of each tumor was examined. The present study was performed in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. The protocol was approved by the Committee on the Ethics of Animal Experiments of the Nanjing Medical University.
Immunohistochemical (IHC) {#S0002-S2008}
-------------------------
The mouse xenograft tumors were immunostained for the marker of Ki-67, E-cadherin and Vimentin.
Statistical Analysis {#S0002-S2009}
--------------------
All the experiments were replicated 3 times. All data were shown as means ± SD. Statistical differences were tested using Student's *t*-test (two-tailed) between two groups and one-way analysis of variance (ANOVA) among multiple groups by SPSS 16.0 software. P≤0.05 was considered to be statistically significant.
Results {#S0003}
=======
SPRY4-IT1 Is Downregulated in DDP-Resistant NSCLC Cells (A549/DDP) {#S0003-S2001}
------------------------------------------------------------------
To evaluate the differences of A549/DDP and A549 cells on DDP treatment response, CCK8 assay was performed to detect the cell viability in the presence of DDP. It showed that A549/DPP cells were more resistant to DDP compared to A549 cells as determined by their IC50 (half-maximal inhibitory concentration) respectively (28.36± 2.3 µg/mL vs 8.75±1.0µg/mL; [Figure 1A](#F0001){ref-type="fig"}; P \< 0.01). A549/DDP cell appears large swellings and spindle-shaped form different from A549 cell under microscopy ([Figure 1B](#F0001){ref-type="fig"}). Considering the role of SPRY4-IT1 in NSCLC development,[@CIT0001] we detected the expression of SPRY4-IT1 in these two cell lines. It was revealed that SPRY4-IT1 expression in A549/DDP cells was significantly decreased by approximately 2.5 folds ([Figure 1C](#F0001){ref-type="fig"}; P\<0.01) in contrast to A549 cells. To further investigate the effects of SPRY4-IT1 on the development of DDP resistance in NSCLC, pCDNA-SPRY4-IT1 vector was transfected into A549/DDP cells to up-regulate SPRY4-IT1 expression ([Figure 1D](#F0001){ref-type="fig"}). CCK8 assay demonstrated that IC50 of SPRY4-IT1 over-expressed cells was largely decreased compared with controls (12.36±1.2µg/mL vs 27.44±2.5µg/mL; [Figure 1E](#F0001){ref-type="fig"}; P \< 0.01).Figure 1lncRNA SPRY4-IT1 expression is downregulated in A549/DDP cells.**Notes:** (**A**) IC50 of cisplatin were detected in A549 and A549/DDP cells. (**B**) Morphologies of A549 and A549/DDP cells under microscope (40×magnification). (**C**) SPRY4-IT1 expression was significantly downregulated in A549/DDP cells compared with A549 cells. (**D**) A549/DDP cells were transfected with pCDNA-SPRY4-IT1 and pCDNA-NC. (**E**) IC50 of cisplatin were detected in A549/DDP cells transfected with pCDNA-SPRY4-IT1 and pCDNA-NC individually. \*\*P\<0.01.**Abbreviations:** lncRNA, long noncoding RNA; SPRY4-IT1, SPRY4 intronic transcript 1; IC50, half-maximal inhibitory concentration; RT-qPCR, quantitative real-time polymerase chain reaction.
SPRY4-IT1 Overexpression Inhibits Proliferation and Promotes Apoptosis in A549/DDP Cells {#S0003-S2002}
----------------------------------------------------------------------------------------
To investigate how SPRY4-IT1 influences the development of NSCLC cell resistance to DDP, the CCK8 assay was conducted. The results of CCK8 assay revealed that cell growth was impaired in A549/DDP cells transfected with pcDNA-SPRY4-IT1 combined with DDP treatment than that in controls ([Figure 2A](#F0002){ref-type="fig"}). To further verify the results, colony formation assay was performed and the results showed that clonogenic survival was inhibited following overexpression of SPRY4-IT1 under DDP treatment (DDP dose of 0.0, 8.0 μg/mL; [Figure 2B](#F0002){ref-type="fig"}). Furthermore, flow cytometry analysis showed that upregulation of SPRY4-IT1 expression significantly promoted DDP-induced apoptosis in A549/DDP cells (DDP dose of 0.0, 8.0 μg/mL; [Figure 2C](#F0002){ref-type="fig"}).Figure 2lncRNA SPRY4-IT1 inhibited cell proliferation and promoted apoptosis induced by cisplatin in A549/DDP cells.**Notes:** (**A**) Cell viability was measured by CCK8 assay in A549 and A549/DDP cells with and without transfection of pCDNA-NC or pCDNA-SPRY4-IT1 under different concentrations of cisplatin after treatment. (**B**) Clonal forming analysis was used to detect the effect of overexpression of SPRY4-IT1 on cell growth under different concentrations of cisplatin at 14 days after treatment. (**C**) Flow cytometric analysis was used to investigate the effect of SPRY4-IT1 on cell apoptosis induced by cisplatin (cisplatin dose of 0.0, 8.0 μg/mL) at 24 h after treatment. \*P \<0.05, \*\*P\<0.01.**Abbreviations:** CCK-8, cell counting kit-8; lncRNA, long noncoding RNA; SPRY4-IT1, SPRY4 intronic transcript 1; RT-qPCR, quantitative real-time polymerase chain reaction.
MPZL-1 Is Negatively Modulated by SPRY4-IT1 in A549/DDP Cells {#S0003-S2003}
-------------------------------------------------------------
RNA transcriptome sequencing was applied to identify genes that exhibited altered expression between SPRY4-IT1 overexpressed cells and controls ([Table 1](#T0001){ref-type="table"}). Ninety-two genes were differentially expressed in all the three replicates (expression change ≥1.5 fold; P\<0.05), and 53 genes were upregulated and 39 genes were downregulated ([Figure 3A](#F0003){ref-type="fig"}). The Gene Ontology (Go) enrichment analysis demonstrated that those genes with significant differences (≥2.5 fold change) following SPRY4-IT1 overexpression were primarily involved in apoptosis, epithelial--mesenchymal transformation (EMT), and mesenchymal cell development pathways ([Figure 3B](#F0003){ref-type="fig"}). To validate the microarray findings, the top 10 up-regulated and down-regulated genes were subjected to verification by RT-qPCR ([Figure 3C](#F0003){ref-type="fig"} and [Table 2](#T0002){ref-type="table"}). It was revealed that the trend in expression changes (upregulation or downregulation) of all 10 selected genes were consistent with the microarray analysis. However, among those differentiated-expression genes, MPZL-1 expression exhibited the most substantial change (down-regulated) and was enriched in the EMT-associated cell migration pathways. The results of the Western blot analysis further confirmed that upregulation of SPRY4-IT1 could lead to decreased MPZL-1 protein expression in A549/DDP cells ([Figure 3D](#F0003){ref-type="fig"} and [E](#F0003){ref-type="fig"}) while neither upregulation nor downregulation of MPZL-1 had no significant effect on the expression of SPRY4-IT1. These results indicated that the expression of MPZL-1 was negatively regulated by SPRY4-IT1.Table 1Number of Regulated Genes in A549/DDP Cells After Overexpression of SPRY4-IT1 as Compared to Negative ControlsNo. of Genes (≥1.5-Fold)Up-Regulated GenesDown-Regulate GenesReplicate 1298255Replicate 2296303Replicate 3347290Replicate 1 + 2 + 35339[^2][^3] Table 2The Dysregulated Gene Profiles After Overexpression of SPRY4-IT1 Using RNA Transcriptome SequencingmRNAsRegulationRatiomRNAsRegulationRatioRP11-166D19.1Up20.1453RP11-1006G14.4Down13.9358RP11-762I7.5Up12.6629NPHP3-ACAD11Down9.39215RESTUp10.4459ADAM1ADown8.47636ZNF322Up9.92374CTBP1-AS2Down4.96036RP11-903H12.5Up6.28244MPZL1Down4.91361ZNF24Up5.47958KIAA0408Down3.69406TUBUp4.29358DNAJB5Down3.32472CSUp3.18511PPAN-P2RY11Down2.90913GSKIPUp2.82117NFAT5Down2.46475SLC16A1Up2.31973TMEM47Down2.13703[^4][^5] Figure 3MPZL-1 was down regulated by SPRY4-IT1.**Notes:** (**A**) RNA transcriptome sequencing was used to detect the dysregulated genes after SPRY4-IT1 overexpression 24 h. (**B**) Gene Ontology (GO) enrichment analysis. (**C**) Ten up-regulated and ten down-regulated genes were identified by RNA transcriptome sequencing. (**D**) MPZL-1 was verified by qRT-PCR to be the target of SPRY4-IT1. (**E**) MPZL-1 was verified by Western blot to be the target of SPRY4-IT1.**Abbreviations:** GO, Gene Ontology; SPRY4-IT1, SPRY4 intronic transcript 1; RT-qPCR, quantitative real-time polymerase chain reaction.
SPRY4-IT1 Reverses Cisplatin Resistance via Downregulating MPZL-1 {#S0003-S2004}
-----------------------------------------------------------------
In order to investigate whether MPZL-1 participates in the development of chemoresistance in A549/DDP cells, we first detected the mRNA and protein expression levels of MPZL-1 in parental A549 and A549/DDP cells. It was confirmed that MPZL-1 expression was upregulated in A549/DDP cells compared with parental A549 cells significantly ([Figure 4A](#F0004){ref-type="fig"}; P\<0.01). Next, it showed that decreased expression of MPZL-1 by siRNA-MPZL-1 transfection ([Figure 4B](#F0004){ref-type="fig"}) inhibited cell proliferation induced by cisplatin, as observed from the survival curve and clonogenic survival assay ([Figure 4C](#F0004){ref-type="fig"} and [D](#F0004){ref-type="fig"}). Furthermore, down-regulation of MPZL-1 promoted A549/DDP cell apoptosis ([Figure 4E](#F0004){ref-type="fig"}). To further verify whether MPZL-1 was involved in SPRY4-IT1-induced regulation of DDP sensitivity, we performed rescue experiments. A549/DDP cells were co-transfected with pCDNA-SPRY4-IT1 and pCDNA-MPZL-1, we found that over-expression of MPZL-1 partially compromised the inhibition effects of SPRY4-IT1 on growth of A549/DDP cells, thereby restoring DDP resistance ([Figure 4F](#F0004){ref-type="fig"}). Our results revealed that the effects of SPRY4-IT1 on NSCLC DDP resistance is at least in part through targeting MPZL-1.Figure 4Knockdown of MPZL-1 inhibited cell growth and induced apoptosis of A549/DDP cells.**Notes:** (**A**) MPZL-1 was upregulated in A549/DDP cells by qRT-PCR and Western blot. (**B**) The expression of MPZL-1 was analyzed by qRT-PCR and Western blot after transfection with si-MPZL-1 in A549/DDP cells at 24 h after treatment. (**C**) Cell growth was inhibited by downregulating MPZL-1. (**D**) Clonal forming analysis was used to detect the effect of MPZL-1 low expression on cell growth. (**E**) Flow cytometric analysis was used to investigate the effect of MPZL-1 low expression on cell apoptosis induced by cisplatin (Cisplatin dose of 0.0, 8.0 μg/mL) at 24 h after treatment. (**F**) Cell growth was measured by CCK8 assay by transfection with pCDNA-SPRY4-IT1 and pCDNA-MPZL-1 contrast to pCDNA-SPRY4-IT1 overexpression group at 0 h, 24 h, 48 h, and 72 h after treatment. \*P \<0.05, \*\*P\<0.01.**Abbreviations:** CCK-8, cell counting kit-8; siRNA, small interfering RNA; RT-qPCR, quantitative real-time polymerase chain reaction; SPRY4-IT1, SPRY4 intronic transcript 1.
SPRY4-IT1/MPZL-1 Affects Epithelial--Mesenchymal Transition in A549/DDP Cells {#S0003-S2005}
-----------------------------------------------------------------------------
Our previous study has documented that SPRY4-IT1 could affect cell epithelial--mesenchymal transition (EMT) process in NSCLC cells, and EMT has been identified as an important effector in the development of chemoresistance. To verify whether SPRY4-IT1/MPZL-1 confer on NSCLC cells DDP resistance via affecting EMT, we detected E-cadherin and Vimentin mRNA and protein expression levels after up-regulation of SPRY4-IT1 or down-regulation of MPZL-1. The results of RT-qPCR ([Figure 5A](#F0005){ref-type="fig"} and [C](#F0005){ref-type="fig"}) and Western blot ([Figure 5D](#F0005){ref-type="fig"} and [B](#F0005){ref-type="fig"}) determined that over-expression of SPRY4-IT1 or down-regulation of MPZL-1 could increase E-cadherin expression while decrease Vimentin expression.Figure 5Upregulation of SPRY4-IT1 and downregulation of MPZL-1 induced EMT reversal.**Notes:** (**A**) The expressions of Vimentin and E-cadherin were detected by qRT-PCR in A549/DDP cells transfected with si-MPZL-1 and NC at 24 h after treatment. (**B**) The expressions of Vimentin and E-cadherin were detected by Western blot in A549/DDP cells transfected with si-MPZL-1 and NC. (**C**) The expressions of Vimentin and E-cadherin were detected by qRT-PCR in A549/DDP cells transfected with pCDNA-SPRY4-IT1 and NC at 24 h after treatment. (**D**) The expressions of Vimentin and E-cadherin were detected by Western blot in A549/DDP cells transfected with pCDNA-SPRY4-IT1 and NC at 24 h after treatment. \*\*P\<0.01.**Abbreviations:** EMT, epithelial--mesenchymal transition; RT-qPCR, quantitative real-time polymerase chain reaction; SPRY4-IT1, SPRY4 intronic transcript 1.
Overexpression of SPRY4-IT1 Inhibits Xenograft Tumor Growth Under DDP Treatment {#S0003-S2006}
-------------------------------------------------------------------------------
In order to validate the effect of SPRY4-IT1 on the sensitivity of A549/DDP cells to DDP in vivo, we constructed nude mouse xenograft model. The results showed that the average tumor volume or weight of SPRT4-IT1 overexpression group was significantly smaller or lighter than that of controls under cisplatin treatment ([Figure 6A](#F0006){ref-type="fig"}--[C](#F0006){ref-type="fig"}). The expression of SPRY4-IT1 in tumor tissues from pCDNA-SPRY4-IT1 transfected A549/DDP cells was significantly increased than controls ([Figure 6D](#F0006){ref-type="fig"}). Moreover, immunohistochemical staining revealed decreased Ki-67 expression, reflecting proliferation inhibition in tumor tissues formed from A549/DDP cells with lncRNA SPRY4-IT1 overexpression ([Figure 6E](#F0006){ref-type="fig"}). In addition, relative expression of epithelial marker (E‑cadherin) was increased while mesenchymal marker (Vimentin) was decreased in pCDNA-SPRY4-IT1 group ([Figure 6E](#F0006){ref-type="fig"}).Figure 6Overexpression of SPRY4-IT1 enhanced cisplatin sensitivity of NSCLC cell in vivo.**Notes:** (**A**) Tumor formed of A549/DDP cells transfected with pCDNA/SPRY4-IT1 after cisplatin treatment was inhibited in vivo. (**B** and **C**) Tumor weight and volume of A549/DDP cells transfected with pCDNA/NC and pCDNA/SPRY4-IT1 after cisplatin treatment in vivo. (**D**) qRT-PCR was used to detect the SPRY4-IT1 expression in pCDNA/NC and pCDNA/SPRY4-IT1 group. (**E**) The E-cadherin, Vimentin and Ki-67 expressions were detected by IHC. \*\*P\<0.01.**Abbreviations:** SPRY4-IT1, SPRY4 intronic transcript 1; RT-qPCR, quantitative real-time polymerase chain reaction; IHC, Immunohistochemistry.
Discussion {#S0004}
==========
For NSCLC patients, chemotherapy remains to be the most basic and successful management; however, drug resistance, either primary or acquired, represents a severe barrier in the improvement of long-term survival and quality of life. With the development of molecular analytical and profiling techniques, the current understanding of resistance mechanisms and certain novel chemoresistance-associated lncRNAs have been identified.[@CIT0011] In the case of NSCLC, multiple lncRNAs, such as CCAT1, AK126698, HOTAIR, GAS5, UCA1 and MEG3 have been reported to influence chemoresistance.[@CIT0012] We previously reported that HOTAIR contributes to cisplatin resistance by downregulating P21[@CIT0013] and MEG3 has effect on mitochondrial apoptosis pathway to lead to cisplatin resistance.[@CIT0014] In this study, it was first revealed that lncRNA SPRY4-IT1 was downregulated in NSCLC cisplatin-resistant cells and up-regulation of SPRY4-IT1 partially reversed DDP resistance of NSCLC cells by promoting apoptosis and inhibiting proliferation.
Despite the detailed molecular mechanisms underlying chemoresistance not yet being fully understood, an accumulating researches have revealed the correlation of drug resistance and EMT.[@CIT0015],[@CIT0016] EMT is a physiopathological process in which epithelial cells lose adhesion proteins and acquire the characteristics of mesenchymal cells.[@CIT0017],[@CIT0018] According to certain clinical data regarding EGFR inhibitors (EGFR-TKIs), it has been suggested that EMT markers may be associated with limited response to EGFR inhibitors, whereas the retention of the epithelial phenotype is correlated with response, even in those without EGFR mutations. A study by Lim in breast cancer demonstrated that EMT confers chemoresistance by regulating the genes associated with cell death and stem cell maintenance. Another study by Pattabiraman revealed that abnormally activated EMT displayed elevated levels of ATP-binding cassette (ABC) that promoted drug efflux.[@CIT0021] In the present study, we observed that forced SPRY4-IT1 expression reversed the cisplatin-resistant phenotype of NSCLC, in large part, owing to its ability to inhibit EMT process.
Further RNA sequencing revealed that MPZL-1 may be a potential SPRY4-IT1 target in NSCLC DDP-resistant cells. In addition, MPZL-1 upregulation had been identified in NSCLC cisplatin-resistant cells (A549/DDP), and down-regulation of MPZL-1 reversed cisplatin resistance and suppressed EMT process. MPZL-1, also known as PZR, has been identified as Src homology phosphatase type-2 (SHP-2) binding partner in epithelial cells which is essential for haematopoietic skeletal and vascular development.[@CIT0019],[@CIT0020] Recent studies have shown that MPZL1 could promote the fibronectin-dependent migration of murine mesenchymal-derived MEF cells, and may be involved in adhesion-dependent signaling.[@CIT0021] Jia et al found that MPZL1 increased the in vitro migratory and in vivo metastatic potential of HCC cells by elevating phosphorylation of ERK-1/2 and AKT, which contribute to EMT programmes via receptor tyrosine kinase (RTK) pathway.[@CIT0022] So, we hypothesized that in NSCLC, MPZL-1 may act as an activator of intracellular signaling pathway regulating EMT, conferring a greater resistance to DDP of NSCLC cells. It has been established that lncRNAs can either act as positive or negative regulators of target gene expression, at various levels, including chromatin modification, transcription, and post-transcriptional, depending on their subcellular location.[@CIT0023],[@CIT0024] Various lncRNAs can recruit special chromatin remodeling complexes to particular chromatin loci to induce epigenetic silence.[@CIT0025] There are also reports on some lncRNAs which regulate the transcription of target genes via a direct mechanism,[@CIT0026] such as promoter and enhancer associated transcripts (paRNAs and eRNAs, respectively), besides, by regulating transcription of enhancers and promoters, they can play negative regulation of gene transcription. Additionally, the involvement of lncRNAs in post-transcriptional modifications can reflect gene function and stability. In summary, our findings first reveal that overexpression SPRY4-IT1 can reverse resistance to cisplatin of A549/DDP cells in vitro and vivo in lung cancers by downregulating MPZL-1 via suppressing EMT. However, the possible mechanisms underlying SPRY4-IT1 negative regulation of MPZL-1 remain to be discovered and need our further work.
Conclusions {#S0005}
===========
SPRY4-IT1/MPZL-1 is a promising target for improving cisplatin efficiency in the treatment of NSCLC.
This work was supported by grants from the National Natural Science Foundation of China (Nos. 81672307, 81501980, and 81871871), the Medical Innovation Team Foundation of the Jiangsu Provincial Enhancement Health Project (No. CXTDA2017021), the "333 high class Talented Man Project" (No. 2016-II-426), Jiangsu Provincial Six Talent Peaks (No. WSW-264), Education and Health Medicine Innovation Team Project (No. CXTDA2017042) and the Nanjing Science and Technology Development Project (No. 2017sc512046).
Disclosure {#S0006}
==========
The authors have no conflicts of interest to declare in this work.
[^1]: These authors contributed equally to this work
[^2]: **Notes:** Expression change of downstream genes (expression change ≥1.5 fold; P\<0.05).
[^3]: **Abbreviation:** SPRY4-IT1, SPRY4 intronic transcript 1.
[^4]: **Notes:** Ten up-regulated genes and ten down-regulated genes.
[^5]: **Abbreviation:** SPRY4-IT1, SPRY4 intronic transcript 1.
|
{
"pile_set_name": "PubMed Central"
}
|
Metabolism of 8-chloro-6-(o-chlorophenyl)-1-methyl-4H-s-triazolo [4,3-alpha] [1,4]benzodiazepine, triazolam, a new central depressant. II. Identification and determination of metabolites in rats and dogs.
1. Eight metabolites of triazolam have been identified, namely, triazolam, dichlorotriazolobenzophenone (DCTB), 1'-hydroxytriazolam, dichloro-alpha-hydroxytriazolobenzophenone (1'-hydroxy-DCTB), Ar-hydroxytriazolam, 4-hydroxytriazolam, Ar-1'-dihydroxytriazolam and 1',4-dihydroxytriazolam. 2. Major metabolites found in the urine were 1',4-dihydroxytriazolam, 1'-hydroxy-DCTB and DCTB in rats; 1'-hydroxytriazolam, 4-hydroxytriazolam and conjugated 1'-hydroxytriazolam in dogs. 3. Major metabolites found in the faeces were 4-hydroxytriazolam in rats; 1'-hydroxytriazolam and 4-hydroxytriazolam in dogs. 4. Conjugated 4-hydroxytriazolam was the major metabolite in both the original and reabsorbed bile of rats. 5. Major metabolites in free form in the plasma were 4-hydroxytriazolam and 1'-hydroxytriazolam in rats; triazolam and 1'-hydroxytriazolam in dogs. 6. The major metabolite in the brain was triazolam, but those in the liver were 4-hydroxytriazolam and triazolam, and in the kidneys were 4-hydroxytriazolam and 1',4-dihydroxytriazolam. 7. Major metabolites in the urine, faeces, plasma and brain after 7-, 14- or 21-day repeated dosing in rats were not much different in type and ratio from those after single dosing. 8. Unchanged triazolam and 1'-hydroxytriazolam were the major metabolites in the plasma, placenta, foetus and amniotic fluid in pregnant rats. 9. There was no change in hepatic aniline hydroxylase and aminopyrine-N-demethylase activity from controls in rats given oral dose of [14C]triazolam for 14 days.
|
{
"pile_set_name": "PubMed Abstracts"
}
|
A milestone many in the medical community would've thought was impossible has been reached at the Neonatal Intensive Care Unit of Jersey City Medical Center.
For five straight years, the unit has sustained zero cases of a healthcare-acquired infection that can be deadly for weak and fragile newborns.
"In general, most units have an infection or a couple infections every thousand line days," said Dr. Ameth Aguirre, director of neonatology and newborn services for JCMC. "We're one of a handful of hospitals in the country with five years, no infections."
Central-line associated bloodstream infections, or CLABSI, occur when bacteria or viruses enter an individual through a central venous catheter, a much-needed tool in order to deliver medicine, fluids or blood products to infants who can't support themselves just yet.
Central lines, different from IVs because of their location, can remain in place for weeks or months, and are therefore more likely to cause a serious infection.
Since 2011, NICUs in nine states — including New Jersey — have participated in a national CLABSI reduction project.
Dr. Kevin Slavin, chief of quality and safety for the Joseph M. Sanzari Children's Hospital at Hackensack University Medical Center, said hospitals have revamped their techniques around the insertion of maintenance of these lines.
"And the nurseries follow strict protocols for removing them as quickly as possible," Slavin said.
The hospital recently reached the year mark with no infant CLABSI cases.
If identified in time, the infection can be reversed. Unlike adults, infants don't present the most obvious symptoms when infected.
"Many thought it would be impossible to get to the point that you can go this long without having a line infection," Slavin said.
Prior to its five-year streak, JCMC's NICU posted a rate higher than the national average, which currently sits around 1.3 infections per 1,000 line days, Dr. Aguirre said. In the five-year CLABSI-free period, the unit has cared for more than 1,500 infants.
|
{
"pile_set_name": "Pile-CC"
}
|
Anti-interleukin-2 receptor antibodies for the prevention of rejection in pediatric renal transplant patients: current status.
The anti-interleukin-2 receptor (anti-IL-2R) antibody therapy is an exciting approach to the prevention of acute rejection after renal allograft transplantation whereby immunosuppression is exerted by a selective and competitive inhibition of IL-2-induced T cell proliferation, a critical pathway of allorecognition. The anti-IL-2R antibodies specifically block the alpha-subunit of the IL-2R on activated T cells, and prevent T cell proliferation and activation of the effector arms of the immune system. The anti-IL-2R antibodies are used as induction therapy, immediately after renal transplantation, for prevention of acute cellular rejection in children and adults. During acute rejection, the IL-2Ralpha chain is no longer expressed on T cells; thus, the antibodies cannot be used to treat an existing acute rejection. Two anti-IL-2R monoclonal antibodies are currently in clinical use: daclizumab and basiliximab. In placebo-controlled phase III clinical trials in adults, daclizumab and basiliximab in combination with calcineurin inhibitor-based immunosuppression, significantly reduced the incidence of acute rejection and corticosteroid-resistant acute rejection without increasing the risk of infectious or malignant complications, and neither antibody was associated with the cytokine-release syndrome. Children who receive calcineurin inhibitors and corticosteroids for maintenance immunosuppression, as well as children who receive augmented immunosuppression to treat acute rejection, are at increased risk of growth impairment, hypertension, hyperlipidemia, lymphoproliferative disorders, diabetes mellitus, and cosmetic changes. In older children, the cosmetic adverse effects frequently reduce compliance with the treatment, and subsequently increase the risk of allograft loss. Being effective and well tolerated in children, the anti-IL-2R antibodies reduce the need for calcineurin inhibitors while maintaining the overall efficacy of the regimen; thus, the anti-IL-2R antibodies increase the safety margin (less toxicity, fewer adverse effects) of the baseline immunosuppression. Secondly, the anti-IL-2R antibodies decrease the need for corticosteroids and muromonab CD3 (OKT3) in children as a result of decreased incidence of acute rejection. The recommended pediatric dose of daclizumab is 1 mg/kg intravenously every 14 days for five doses, with the first dose administered within 24 hours pre-transplantation. This administration regimen maintains daclizumab levels necessary to completely saturate the IL-2Ralpha (5-10 microg/mL) in children for at least 12 weeks.The recommended pediatric dose of basiliximab for recipients <35 kg is 10 mg, and 20 mg for recipients > or =35 kg, intravenously on days 0 and 4 post-transplantation. This administration regimen maintains basiliximab levels necessary to completely saturate the IL-2Ralpha (>0.2 microg/mL) in children for at least 3 weeks.
|
{
"pile_set_name": "PubMed Abstracts"
}
|
Hurricane Sandy last year pushed the issue of climate change higher on the nation’s agenda. President Barack Obama indicated in his inaugural address and his State of the Union address that climate policy would be a priority for his second term. Some members of Congress said Sandy might cause Congress to redesign infrastructure spending to limit damage from future catastrophic storms.
A series of Senate votes Friday indicated what the political balance now is on energy policy and on measures to avert climate change.
These were non-binding votes on a budget resolution that’s almost surely not going to be passed by both chambers of Congress.
But they show where senators stand, with a third of them up for election next year and presumably focused on the political impacts of their policy choices. Damaging campaign ads could be produced from a vote on any amendment that put a senator at odds with public opinion in his or her state.
One interpretation of the votes: a bipartisan majority of senators do not want to impose a fee or tax on carbon dioxide emissions – but also do not want to stop the Environmental Protection Agency from regulating greenhouse gas emissions. Yet a majority also wants to increase the supply of energy even if that energy would come from Canadian oil sands that generate more greenhouse gas emission than other grades of oil.
Taken together the votes seem to be further evidence that in the near term the EPA, not Congress, will be the primary venue for the legal and policy battles over climate change. Obama’s nominee to head the EPA will have her confirmation hearing next month before the Senate Environment and Public Works Committee.
Alex Wong / Getty Images Sen. Sheldon Whitehouse, D-R.I., and Rep. Henry Waxman, D-Calif., speak to the media during a news conference January 24, 2013 on Capitol Hill.
On the carbon fee or tax, Sen. Sheldon Whitehouse, D-R.I., proposed “the establishment of a fee on carbon pollution,” with the provision that all revenue from such a fee would be “returned to the American people in the form of Federal deficit reduction, reduced Federal tax rates, cost savings, or other direct benefits.”
“We ignore carbon pollution at our peril, and we have subsidized it long enough,” Whitehouse said before the vote on his amendment. “It is past time to wake up from our sleepwalking. This vote is a test. Whether we pass or fail is a measure of us.”
By a vote of 58 to 41, Whitehouse’s fee was defeated. No Republican senators voted for it and 13 Democratic senators – including six Democrats up for re-election next year – voted against it.
In a separate vote the Senate rejected an amendment by Sen. Roy Blunt, R-Mo., that would have made it more difficult in the future to enact a carbon tax or fee, by establishing what’s called a “a budget point of order” against such a tax.
Blunt got 53 votes; he needed 60. Six Democrats, including three up for re-election in 2014 -- Sens. Max Baucus of Montana, Mary Landrieu of Louisiana and Mark Pryor of Arkansas -- voted for Blunt’s amendment.
Franz Matzner, associate director of government affairs for the Natural Resources Defense Council, an environmental advocacy group, said the vote on the Whitehouse and Blunt amendments indicated that senators were “not ready to take it (a carbon tax) off the table and say that we going to make it so that in the future no (carbon tax) legislation can be considered because it will violate the terms of the budget.”
The Senate also voted, 62 to 37, to support the building of the Keystone XL pipeline project which would connect oil sands production facilities in Alberta, Canada with refineries in the United States.
Seventeen Democratic senators – including eight Democrats up for re-election next year – voted for an amendment proposed by Sen. John Hoeven, R-N.D., expressing support for building the pipeline.
The decision right now rests in the hands of the Obama administration. Under an executive order dating back to 1968, the secretary of state decides on whether to permit a pipeline, such as Keystone XL, that connects the United States with a foreign country.
Hoeven said approving the pipeline is matter of “getting our economy going and growing, and… making sure we don’t have to import oil from the Middle East. It is not just oil from Canada, it is oil from the great State of North Dakota and Montana — light, sweet crude we need to get to our refineries.”
But Sen. Barbara Boxer, chairman of the Environment and Public Works Committee, said the Senate must first grapple with questions such as whether the increased use of oil sands from Canada will lead to higher greenhouse gas (GHG) emissions.
A Congressional Research Service report last week said that Canadian oil sands crudes “are on average somewhat more GHG emission intensive than the crudes they may displace in U.S. refineries,” up to 20 percent higher than for the average transportation fuel.
Boxer offered an amendment – in contrast with Hoeven’s -- that she said would ensure important issues would be addressed, “such as how much oil will stay here versus how much will be exported and, therefore, will we suffer from higher energy prices? How much steel will be made in America?” She added that, “Our American national security experts warn us against the instability worldwide caused by climate disruption.” Boxer’s amendment was defeated, 58 to 41.
In another big environmental vote, the Senate, 52 to 47, defeated an amended by Sen. Jim Inhofe, R- Okla. that in Inhofe’s own words “stops the EPA from having the jurisdiction over the regulation of carbon” and would have defunded the agency’s greenhouse gas regulations.
On this vote, only three Democrats, Pryor, Landrieu, and Sen. Joe Manchin of West Virginia voted for Inhofe’s amendment. Sen. Susan Collins of Maine, who is up for re-election in 2014, was the only Republican senator voting against Inhofe’s amendment.
“The good news is that majority of the Senate clearly acknowledges that climate change is a reality that has to be addressed and they didn’t want to tie the hands of Congress or the administration to find ways to address carbon pollution,” Matzner said.
|
{
"pile_set_name": "OpenWebText2"
}
|
Grubby Profile Joined January 2011 Netherlands 318 Posts Last Edited: 2013-01-05 09:49:56 #1
I'll list some interesting trivia of facts which might be little known to the foreign fans as I find them out! Will keep updating during my stay here.
They pronounce FXOTear like you would a tear from the eye .
. FXOTear says there is no meaning to his name, neither "tear as in to rip" nor the "tear from the eye". That answer didn't satisfy, so I quizzed, "if you had to pick a meaning, which one would you go for?". He gesticulated to rip (probably his enemies).
(probably his enemies). From the FXO.kr members, JkS has the best English
Leenock and Kong are among the youngest members
Choya and Supreme are functioning as coaches
I am the oldest one in the house right now lol (1986)
On January 04 2013 18:38 Wafflelisk wrote:
What do they all like to eat for breakfast? I find myself lacking knowledge of progamer diets =(
This: http://t.co/HxkaGX2u
On January 04 2013 18:40 Torte de Lini wrote:
Whats the practice routine like?
10am start, 2 am finish. Two 2-hour breaks for food. Sometimes people relax a bit more than the strict routine, sometimes they are all playing (I'm not fully aware what exceptions exist. For example, are there free days in a week etc?) On January 04 2013 19:36 Fragile51 wrote:
Does the team do anything physical, like go on jogs/the gym or something similar to keep in good physical condition? I remember slayers always used to play soccer every so often.
A bunch of them go to the gym, some (like Lucky) 6x a week. They'll do 50% cardio at least or more, and some weight training for example for chest or arms (at least from what I saw today, only went along once so far). They play soccer in spring too, apparently they're the #2 team in soccer only losing to MVP haha.
On January 04 2013 20:11 GreyStar wrote:
So they have a 10am-2pm practice requirement ?
No it's a little looser than that. Some people go to bed earlier, while others wake up or start later. It's not as strict as I made it sound, these are just the hours people generally practice in. No one can really reach that and feel healthy for a long period of time I don't think.
NEW UPDATES 05 January:
On January 05 2013 02:49 bram--1995 wrote:
what do the players do in there spare time?
They socialize in other parts of the house, a handful go fitness together. Some read anime's on the PC or watch Korean drama shows. Basically the Korean equivalent of what Western hardcore gamers would also do in their off time.
On January 05 2013 03:13 mcmartini wrote:
Who has impressed you the most in terms of gameplay and hand speed?
Because he is also Protoss (my race), that's FXOTear. The house immediately said he's their Top protoss, and he's very fast. He's also got some very cool builds which have been really inspiring in general.
On January 05 2013 03:50 frequency wrote:
JKS has the best English? I always thought his English was good but he seemed to scared to type it :O
JkS previously said his speaking is better than his writing due to hands-on experience.
On January 05 2013 03:50 Whitewing wrote:
Who is the best singer on the team?
Well I'm sitting right behind Choya and I can confirm it's not Choya for sure. The others have been shy to sing. Should I check if I can get them drunk, go to a Karaoke bar, put on some Kelly Clarkson and tape the whole thing?
On January 05 2013 18:17 JOJOsc2news wrote:
This is great. Thanks for doing this.
Questions:
How is the atmosphere in the house?
On a scale of OR (operating room) to LAN party, where would you place it?
Good luck!!!
Solidly in the middle; a jovial opera room.
On January 05 2013 18:30 Oboeman wrote:
can you elaborate on the roles of the coaches, and what kind of work they do with the players?
Show nested quote +
Strangely, i read all of that in your voice... LOL
me too, and i even visualized him speaking into a webcam as if it were a vlog.
spooky. can you elaborate on the roles of the coaches, and what kind of work they do with the players?me too, and i even visualized him speaking into a webcam as if it were a vlog.spooky.
Think gentle adjustments. Nudge to wake up, directions on how to practice, showing builds, etc. It helps that Choya is the kind of thinker-Protoss just like Squirtle for example.
I'm staying to train at the FXOpen Korea team house for a period of time, and so far everyone's super nice and I'm happy to be staying here. I'd like to show help them back by getting them a bit more known to the foreign scene.I'll list some interesting trivia of facts which might be little known to the foreign fans as I find them out! Will keep updating during my stay here. Homepage: followgrubby.com Twitter: @followgrubby Facebook: /followgrubby
|
{
"pile_set_name": "OpenWebText2"
}
|
Signing Daniel Alfredsson was the headline-grabbing move of the Detroit Red Wings’ summer, but it wasn’t the biggest one. Instead, that came in an intriguing gamble: handing former Florida Panthers forward Stephen Weiss a five-year, $24.5 million contract to essentially replace Valtteri Filpulla.
In an ideal world, Weiss will fit in seamlessly to the Detroit machine. He’s a versatile forward whose two-way play often got lost in the obscurity that comes with playing in Florida. (Weiss even fits in with an aging core, as he’s logged a lot of mileage at 30.)
Weiss’ contract year was anything but ideal, though.
He was limited just 13 games and the results were pitiful; he scored just four points and registered a -13 rating, needing four contests to register his first shot on goal in 2013.
Beyond the awful run, his injury concerns are significant worry for a player signing a substantial contract.
On the bright side, he’s the type of free agent acquisition who isn’t striving to become his new team’s best player. As Weiss told the National Post, it seems that his goal is merely to fit right in.
“There really wasn’t a lot of thought going into [the deal with Detroit],” Weiss said. “To join an organization like this is pretty special for me. [Yzerman] was one of my favourite players, no doubt. He’s a guy I looked up to for sure. But I’m not dreaming about trying to go in there and doing something similar. Hopefully I can just help out.”
Minnesota Wild goalie Devan Dubnyk has been the most difficult goalies to score against this season. Leave it to a high-level player like Leon Draisaitl to make it look this, well, “easy.”
Draisaitl scored his 13th goal of 2016-17 by capping this pretty give-and-go play with Benoit Pouliot. You can see the frustration from Dubnyk at the end of the tally, as if he was saying “How was I supposed to stop that?” (though probably with more colorful language).
Draisaitl came into Friday with five goals and three assists in his last five games, so he’s been almost unstoppable lately.
|
{
"pile_set_name": "Pile-CC"
}
|
Evaluation of western fence lizards (Sceloporus occidentalis) and eastern fence lizards (Sceloporus undulatus) as laboratory reptile models for toxicological investigations.
A need is recognized for one or more laboratory reptile models for use in ecotoxicological studies and risk assessments. Maintenance of breeding populations of most reptile species under laboratory conditions is not practical because of their size and slow maturation rate. However, a number of species of spiny lizards (Sceloporus sp.) are small, mature quickly, and reproduce under laboratory conditions. We evaluated three populations of western fence lizards (S. occidentalis) and four populations of eastern fence lizards (S. undulatus) for their performance under laboratory conditions. We reared an F1 generation of each population and compared their performance relative to survival, growth, maturation rate, and reproductive output. A population from the San Joaquin Valley (CA. USA) performed especially well under laboratory conditions and is a good candidate for a laboratory model. We also examined the sensitivity of developing fence lizard embryos to an estrogenic chemical to determine if male secondary sex characteristics were affected. Microinjecting eggs with an estrogenic chemical (17alpha-ethinylestradiol) feminized males and prevented development of embryonic secondary sex characteristics. Therefore, embryonic fence lizards may be useful for studying the effects of endocrine-disrupting chemicals.
|
{
"pile_set_name": "PubMed Abstracts"
}
|
The securing of data during transmission has been of interest throughout human history. Secure communication has been an essential part of commerce since time immemorial.
More recently, and especially since the widespread availability of computing power and technical means of data transmission, with sophisticated means of securing data transmitted over telecommunications channels and equally sophisticated technical means of decrypting messages, there has developed a rapidly-accelerating race between those who wish to secure messages and those who wish to “crack” them. There is a constant search for new technical means of securing data during transmission by increasing the threshold of feasibility of decryption, and an equally constant search for means of rendering feasible decryptions that were thought to be infeasible. Similarly, the processing and transmission costs of sending information securely are of concern. The volume of data to be transmitted in the course of business transactions is increasing, and the cost of using public networks is constantly decreasing, while the cost of using private networks is ever more costly. It would be advantageous to be able to send more data, especially in bulk data applications, over less costly open channels, such as the Internet, but it is difficult to secure transmissions over such a medium to the standard normally required for commercial confidentiality purposes.
There have been attempts to alleviate the problem of combining security with low cost.
Published European patent application number EP 0 993 142 A1, for example, proposes a method for providing security for data wherein the bulk of transmitted data is encrypted and transmitted over an inherently less secure channel while selected segments of data are transmitted over a normally private channel, such as the telephone network. An eavesdropper on the less secure channel is thus prevented from reading all the data. Disclosed also is the notion of using one or more scrambling algorithms to scramble data according to a formula derived from the data itself.
Published PCT patent application number WO 00/18078 proposes a method whereby a message is split and transmitted over two channels in such a manner that the portion of the message to be sent over the less secure channel is encrypted, while the portion transmitted over the secure channel remains unencrypted.
It is desirable to find a way of further increasing the security of a message by reducing the computational feasibility of an unauthorized person's recovering the information content of the message and reducing the cost of processing and transmission.
|
{
"pile_set_name": "USPTO Backgrounds"
}
|
As electronic circuits and systems are becoming ever more popular in critical applications, including, but not limited to, data storage, electronic vehicles and aeronautics, so do their operating conditions become harsher. This often means using electronic components at or near the limit of their functional parameters and very close to or even exceeding their maximum operating temperature. Accurately monitoring the temperature of these electronic circuits and systems therefore becomes critical to avoiding failures or other reliability issues.
In certain temperature measurement applications, such as, for example, monitoring the temperature of a hard disk drive so as to change temperature-sensitive parameters relating thereto, a thermistor is often placed at the disk drive head. A thermistor is a well-known type of device having a resistance that varies significantly with temperature, more so than in standard resistors. A voltage across the thermistor is measured and a corresponding temperature is obtained (e.g., based on a prescribed relationship between temperature and resistance associated with the device) as a function of the voltage across the thermistor.
There are many sources of error using this standard approach. Two primary sources of error are the variation in the external voltage supply, which is typically used to bias the thermistor, and the accuracy of an internal voltage reference often used for comparison with the thermistor voltage. To reduce these errors, a calibration procedure can be employed, such as during power-up. However, calibration increases circuit complexity and must generally be performed by the user, and is therefore undesirable. Furthermore, because calibration is typically only performed once (e.g., at power-up), such an approach cannot correct errors resulting from variations in temperature or other operating conditions which occur after calibration has completed.
|
{
"pile_set_name": "USPTO Backgrounds"
}
|
Synthesis, pairing, and cellular uptake properties of C(6')-functionalized tricyclo-DNA.
Tricyclo-DNA (tc-DNA) is a promising candidate for oligonucleotide-based therapeutic applications exhibiting increased affinity to RNA and increased resistance to nucleases. However, as many other oligonucleotide analogs, tc-DNA does not readily cross cell membranes. We wished to address this issue by preparing a prodrug of tc-DNA containing a metabolically labile group at C(6') that promotes cellular uptake. Two monomeric nucleoside building blocks bearing an ester function at C(6') (tc(ee)-T and tc(hd)-T) were synthesized starting from a known C(6') functionalized bicyclic sugar unit to which the cyclopropane ring was introduced via carbene addition. NIS-mediated nucleosidation of the corresponding glycal with in situ persilylated thymine afforded the β-iodonucleoside exclusively that was dehalogenated via radical reduction. Diversity in the ester function was obtained by hydrolysis and reesterification. The two nucleosides were subsequently incorporated into DNA or tc-DNA by standard phosphoramidite chemistry. The reactivity of the ester function during oligonucleotide deprotection was explored and the corresponding C(6') amide, carboxylic acid, or unchanged ester functions were obtained, depending on the deprotection conditions. Compared to unmodified DNA, these tc-DNA derivatives increased the stability of duplexes investigated with ΔT(m)/mod of +0.4 to +2.0 °C. The only destabilizing residue was tc(hd)-T, most likely due to self-aggregation of the lipophilic side chains in the single stranded oligonucleotide. A decamer containing five tc(hd)-T residues was readily taken up by HeLa and HEK 293T cells without the use of a transfection agent.
|
{
"pile_set_name": "PubMed Abstracts"
}
|
Q:
Percentages in minmax CSS grid causing overflow
I'm trying to build a responsive CSS grid with three columns that wrap when necessary. I want the minimum column width to be 200px, otherwise use the percentage 30%.
I believe I can use the minmax CSS calculation to do this, but some strange overflow error occurs when I use percentages rather than pixels.
.wrapper-pix {
display: grid;
grid-gap: 20px;
grid-template-columns: repeat(auto-fit, minmax(200px, 360px));
}
.wrapper-percent {
display: grid;
grid-gap: 20px;
grid-template-columns: repeat(auto-fit, minmax(200px, 30%));
}
.box-pix{
background: #4299E1;
height: 200px;
}
.box-percent{
background: #F56565;
height: 200px;
}
<div class="wrapper-pix">
<div class="box-pix"></div>
<div class="box-pix"></div>
<div class="box-pix"></div>
</div>
<div class="wrapper-percent">
<div class="box-percent"></div>
<div class="box-percent"></div>
<div class="box-percent"></div>
</div>
Please also see this CodePen: https://codepen.io/willjonty/pen/ExVpVza.
The blue boxes prefer to wrap and stay at their maximum percentage until the window becomes particularly small. The red boxes, however, will resize their max value with the percentage until they reach the 200px breakpoint. So far, this is all good.
I would expect the red boxes to wrap when the viewport reaches 640px, which is 200 * 3 (for the three columns), plus 2 * 20px gaps. Notice however that when the viewport becomes smaller, the red boxes do not wrap at all, and instead overflow into a hidden part of the window. For some reason, the third box will only wrap when the viewport is 399 or less.
Does anyone know why CSS grid behaves like this? And how I might implement a column max-width of 30% that will wrap without any overflow?
I would use fr units, but those will hijack auto-fit to increase the number of columns in a row depending on the data. I want 3 equally-sized columns.
Thank you in advance.
A:
Setting the max-width to a percentage negates the auto-fit declaration in a sense. While fr deals with free space, percentages are an actual width in the browser's eyes. So when you set the max to 30%, the browser will honor that without wrapping, because it can always split whatever width it has into chunks of 30%. This of course, is down to your 200px minimum that you set.
Since you don't want to use fr in your desktop rendering, I would recommend inserting a media query at the breakpoint which your content begins to not work, and reduce to one column that takes up all the free space. Example:
@media only screen and (max-width: 640px) {
.wrapper-percent {
grid-template-columns: 1fr;
}
}
|
{
"pile_set_name": "StackExchange"
}
|
Q:
How to use a Seafile generated upload-link w/o authentication token from command line
With Seafile one is able to create a public upload link (e.g. https://cloud.seafile.com/u/d/98233edf89/) to upload files via Browser w/o authentication.
Seafile webapi does not support any upload w/o authentication token.
How can I use such kind of link from command line with curl or from python script?
A:
needed 2 hours to find a solution with curl, it needs two steps:
make a get-request to the public uplink url with the repo-id as query parameter as follows:
curl 'https://cloud.seafile.com/ajax/u/d/98233edf89/upload/?r=f3e30b25-aad7-4e92-b6fd-4665760dd6f5' -H 'Accept: application/json' -H 'X-Requested-With: XMLHttpRequest'
The answer is (json) a id-link to use in next upload-post e.g.:
{"url": "https://cloud.seafile.com/seafhttp/upload-aj/c2b6d367-22e4-4819-a5fb-6a8f9d783680"}
Use this link to initiate the upload post:
curl 'https://cloud.seafile.com/seafhttp/upload-aj/c2b6d367-22e4-4819-a5fb-6a8f9d783680' -F file=@./tmp/index.html -F filename=index.html -F parent_dir="/my-repo-dir/"
The answer is json again, e.g.
[{"name": "index.html", "id": "0a0742facf24226a2901d258a1c95e369210bcf3", "size": 10521}]
done ;)
|
{
"pile_set_name": "StackExchange"
}
|
1.. Introduction
================
Measuring heterogeneity in satellite imagery is important, since heterogeneity in an image represents the degree of diversity of objects reflecting within a landscape. In fact, since the IFOV (Instantaneous Field of View) of an image represents a spatially implicit representation of reality, each pixel is expected to represent reality at a certain resolution.
Despite the attribute being considered, the diversity of that attribute has been proven to change as a function of scale \[[@b1-sensors-09-00303]\]. Most measures of spectral diversity have been proposed based on the Boltzmann index \[[@b2-sensors-09-00303]-[@b3-sensors-09-00303]\], commonly referred to as Shannon entropy index \[[@b4-sensors-09-00303]-[@b6-sensors-09-00303]\] *H* = −Σ*p* × ln(*p*), where *p* is the relative abundance of each spectral reflectance value (Digital Number, DN). The Shannon index will increase if the DN values are equally distributed with no DN value being dominant with respect to the others. The Shannon index has been advocated as a powerful algorithm for measuring diversity. Nonetheless, it does not explicitly consider how the measure of diversity changes as a function of scale if it is applied to an entire image. It may be made to account for the variation of diversity across spatial scales if it is repeatedly calculated while increasing the sampling extent within the chosen study area. This process may be time expensive. Quoting Gorelick \[[@b3-sensors-09-00303]\], who made a critique on diversity measured by Shannon and Simpson indices, one can never capture all aspects of diversity in a single statistic. This is true regardless of the attribute being considered.
The aim of this paper is to propose a method for measuring spectral heterogeneity at multiple scales simultaneously based on ecological theory.
2.. Algorithmic foundation of spectral rarefaction
==================================================
In ecology, there is a long history of dealing with species diversity over space or time. In particular, given *N* plots, i.e. sampling units with a certain dimension, three different kinds of species diversity may be recognized: alpha or local diversity (α), i.e. the number of species within one plotgamma or total diversity (γ), i.e. the number of species considering *N* plotsbeta or between-plots diversity (β), i.e. the diversity deriving from the complementarity of the species composition considering pairs of plots \[[@b1-sensors-09-00303]\].
In this view, accumulation curves, showing the number of accumulated species given a certain number of sampled plots, have long been used for estimating the expected number of species within a study area given a specific sampling effort. Since the order that samples are added to an accumulation curve accounts for its shape \[[@b7-sensors-09-00303]-[@b8-sensors-09-00303]\], an order-free curve is derived by means of (i) an analytical solution or of (ii) permutations of samples \[[@b9-sensors-09-00303]-[@b10-sensors-09-00303]\]. This order-free curve is referred to as a rarefaction curve. Considering permutation (ii), once *N* plots have been visited across a study area and the presence of all species has been recorded (obtaining a presence/absence matrix **M~s~** of *N* plots per *S* species), a rarefaction curve is then obtained by repeatedly resampling the pool of *N* plots at random without replacement and plotting the average number of species represented by *1, 2, ..., N* plots \[[@b6-sensors-09-00303],[@b10-sensors-09-00303]\]. Thus, sample-based rarefaction generates the expected number of accumulated species as the number of sampled plots increases from 1 to *N*.
On the other hand, an analytical solution (i) may be formalized as:
Let **M~s~** be a presence/absence matrix of *N* plots per *S* species, the formal estimate of the expected number of species per number of plots turns out to be: $$E\lbrack S\rbrack = S - \frac{\underset{i = 1}{\overset{S}{\text{∑}}}\left( \begin{array}{l}
{N - N_{i}} \\
n \\
\end{array} \right)}{\left( \begin{array}{l}
N \\
n \\
\end{array} \right)}$$where *N~i~* = number of plots where species *i* is found and *n* = number of randomly chosen plots \[[@b9-sensors-09-00303]-[@b12-sensors-09-00303]\]
Generally, the steeper the curve, the greater the increase in species richness as the sample size increases \[[@b7-sensors-09-00303]-[@b12-sensors-09-00303]\].
From a landscape perspective, rarefaction curves are directly related to the environmental heterogeneity of the area sampled. In fact, it is expected that the greater the landscape heterogeneity, the greater the species diversity, including both fine-scale and coarse-scale species richness (i.e. α- and γ-diversity, respectively), and compositional variability, or β-diversity \[[@b7-sensors-09-00303]\].
Computing β-diversity deals with looking at the difference between pairs of plots in terms of species composition \[[@b13-sensors-09-00303]-[@b15-sensors-09-00303]\]. Popular indices of β-diversity, e.g. the Jaccard index, are based on the intersection of the composition in species between pairs of plots with respect to their union, as *C~j~* = 1 − (∩/∪) \[[@b13-sensors-09-00303]\]. The higher the intersection in species composition the lower the β-diversity. As an example, given two plots with α = 5 species, using *C~j~*, β-diversity will range from β=0 when the 5 species will be exactly the same, while the maximum β-diversity (β=1) will occur when all the 5 species will be different.
An alternative definition of β-diversity has been provided by Whittaker \[[@b1-sensors-09-00303]\] who expressed it as *β* = *γ/ᾱ*. This was later modified by Lande \[[@b16-sensors-09-00303]\] being more consistent with the rarefaction theory. Using rarefaction curves, diversity may be partitioned by additive partitioning as *γ* = *α* + *β* \[[@b16-sensors-09-00303]-[@b17-sensors-09-00303]\], leading to considering β in the same unit of measurement (i.e. number of species) of α and γ as: *β* = *γ* − *α* ([Fig 1](#f1-sensors-09-00303){ref-type="fig"}).
In this paper, different "species" will be replaced by different "DNs" (Digital Numbers, i.e. spectral values).
Consider a satellite image with a radiometric resolution of 8 bit. This means that the reflectance values of the pixels, i.e. the Digital Numbers (DNs), may range from 0 to 255.
Subsampling the image by means of *N* plots, i.e. spatial windows with a certain dimension, will lead to a presence/absence matrix **M~DN~** of *N* plots per *S* DNs.
Given the matrix **M~DN~**, [Eq.(1)](#FD1){ref-type="disp-formula"} previously introduced for species diversity can also provide a formal estimate of the number of DNs per number of windows when *N~i~* = number of plots where the DN value *i* is found.
Therefore, the same concepts introduced for species diversity may thus be applied to satellite imagery diversity. Applying rarefaction theory to DNs rather than species leads to consider three different components of pixels diversity: alpha or local diversity (α~DN~), i.e. the number of different DNs within one plotgamma or total diversity (γ~DN~), i.e. the number of different DNs considering *N* plotsbeta or between-plots diversity (β~DN~), i.e. the diversity deriving from *β*~DN~ = *γ*~DN~ − *α*~DN~ \[[@b16-sensors-09-00303]-[@b17-sensors-09-00303]\].
3.. Worked example
==================
[Eq.(1)](#FD1){ref-type="disp-formula"} only works with one-dimensional systems. In fact the dimension Dim(**M~DN~**) of the presence absence matrix **M~DN~** of *N* plots per *S* DN values equals Dim(**M~DN~**)=(*N,S*), implying that: the plots are rowsthe DN values are columnsthe cells composing the matrix are presence/absence values, i.e. they are dummy coded as 1s and 0s.
For instance, [Fig. 2](#f2-sensors-09-00303){ref-type="fig"} shows the presence/absence matrix **M~DN~** of *N* plots per *S* DN values derived from an 8-bit image sampled by 6 plots, where Dim(**M~DN~**)=(*N,S*)=(6,256), with DN values in one dimension ranging from 0 to 255.
Thus, before building rarefaction curves one should choose a single band to work with. Following biological theory, an infrared waveband should be used when working with vegetation based on its intrinsic capability of discriminating different vegetation types \[[@b18-sensors-09-00303]\]. Another option may be based on performing data reduction with a method such as PCA and further using the first principal component explaining most of the variance.
Once the rarefaction algorithm ([Eq.(1)](#FD1){ref-type="disp-formula"}) has been applied to the presence absence matrix **M~DN~**, different study areas sampled by the same number of plots containing the same number of inner pixels (e.g. 1000 pixels per spatial window) will possibly show very different curves. [Fig. 3](#f3-sensors-09-00303){ref-type="fig"} shows two areas with different levels of heterogeneity, each sampled by six spatial windows (plots).
Considering the ecologically heterogeneous area (upper curve of [Fig. 3](#f3-sensors-09-00303){ref-type="fig"}) with respect to the more homogeneous one (lower curve of [Fig. 3](#f3-sensors-09-00303){ref-type="fig"}), α~DN~ equals 55 and 24 respectively, i.e. there are on average 55 and 24 distinct reflectance values for each plot (spatial window).
Meanwhile γ turns out to be 253 and 50, for heterogeneous vs. homogeneous area, respectively.
This means that the spectral value diversity β~DN~ as calculated by γ~DN~-α~DN~ is 198 and 26, respectively.
Notice that in this worked example, the rarefaction algorithm ([Eq.(1)](#FD1){ref-type="disp-formula"}) allowed us to: (i) represent the variation in diversity by means of a single algorithm applied to a matrix, i.e. the spatial variation of diversity components, (ii) represent β-diversity by means of only one statistic without considering pair-wise distances among spatial windows (as in the case of e.g. the Jaccard index), and (iii) represent β-diversity in the same units as α- and γ- diversity (i.e. number of different DN values).
4.. Remarks and summary
=======================
Rarefaction and additive partitioning of diversity, which are often used in ecology with reference to species diversity \[[@b6-sensors-09-00303]-[@b12-sensors-09-00303],[@b16-sensors-09-00303]-[@b17-sensors-09-00303]\], could be applied to reflectance values for estimating and graphically representing local (α), global (γ) and turnover in (β) environmental variability. In fact, once rarefaction curves are graphed, it becomes apparent that α, γ and β represent the minimum, the maximum and the turnover (i.e. maximum - minimum) of the curve ([Fig. 1](#f1-sensors-09-00303){ref-type="fig"}). Generally speaking, it is expected that the higher the minimum value the higher the local variability within a plot, while, given the same local variability, the higher the slope of the curve the higher the variability across the different plots within the area \[[@b7-sensors-09-00303],[@b10-sensors-09-00303],[@b19-sensors-09-00303]\]. On the contrary if the slope is low, i.e. when the curve rapidly reaches the asymptote, the accumulated spectral values are simply a replicate of the sampled spectral values, thus indicating global homogeneity of the area.
In summary, for each plot (spatial window), containing a number *n* of pixels (e.g. 1000 pixels per window), the number of different DNs should theoretically range from 1 (homogeneous environment such as water) to 256 (heterogeneous environment composed of different land cover classes, with a 8-bit image). Notice that a lower maximum number of DNs per plot (window) is expected on the strength of the spatial autocorrelation of spectral values. Once spectral rarefaction curves are built, the number of DN values per window is directly estimated (α) and rises until theoretically reaching the maximum value of 256 (in case of commonly used 8-bit images) as new plots are added to the curve. Obviously the theoretical maximum of 256 different values is reached only when the considered area is so heterogeneous that it comprises all the 256 values.
The approach proposed for measuring spectral heterogeneity is robust but straightforward and consists of three main tasks: (i) selecting within the image adjacent or random windows containing a given number of pixels; (ii) choosing one band ([Eq.(1)](#FD1){ref-type="disp-formula"} only works with one-dimensional systems, see section "2. Worked example"), (iii) performing rarefaction curves by [Eq.(1)](#FD1){ref-type="disp-formula"} and estimating α-, β− and γ- diversity components.
Of course, other techniques rather than spectral rarefaction could account for the spatial variability of DN values as well, e.g. semivariograms \[[@b14-sensors-09-00303],[@b20-sensors-09-00303]\]. However, spectral rarefaction coupled with additive partitioning exhibits mathematical and statistical properties which may be directly related to spectral and species α-, β- and γ- diversity. This is an enormous advantage to using spectral rarefaction as a straightforward method for (i) robustly estimating local to global diversity of an area directly relating sensor-based and field-based heterogeneity and (ii) quantitatively comparing different areas with different degrees of heterogeneity at multiple scales.
I strongly acknowledge Root Gorelick, Carlo Ricotta and a third anonymous referee for precious insights on a previous draft of the paper. I am particularly grateful to Brian S. Cade and Daniel J. McGlinn for additional comments on the manuscript.
{#f1-sensors-09-00303}
{#f2-sensors-09-00303}
{ref-type="disp-formula"} provide an estimate of the number of different DNs at various spatial scales. Obviously only one band or the first PC can be used at once. See the main text for major explanations.](sensors-09-00303f3){#f3-sensors-09-00303}
|
{
"pile_set_name": "PubMed Central"
}
|
ICM Registry
ICM Registry operates the .xxx (pronounced "dot triple-X") sponsored top-level domain (sTLD) registry, which is designed for pornography. The ICM Registry operates from Palm Beach Gardens, Florida. It is owned by Stuart Lawley.
History
In 2005, the Bush Administration pressured ICANN not to adopt a .xxx rating on ideological grounds.
On 18 March 2011, the ICANN Board voted to approve the .xxx sTLD, which later went into operation on 15 April 2011.
On 12 April 2012, the ICM Registry announced their applications for additional sTLDs .SEX, .PORN and .ADULT.
References
Category:Web accessibility
Category:Domain registries
Category:Companies based in Palm Beach County, Florida
Category:Palm Beach Gardens, Florida
Category:Companies with year of establishment missing
Category:Sexuality and computing
|
{
"pile_set_name": "Wikipedia (en)"
}
|
My inlaws have asked my opinion on a pool table and I have nothing to tell them. Has anyone heard of Golden West Billiard Mfg? The table they were looking at was from the Classic line called the Sunrise. This is a table that Mike Sigel is selling in his retail store but I have never heard of it before. The website is http://www.billiardmfg.com/home.asp
If you have heard of the table, how would you compare it to a Proline. The two tables they are looking at are 4' x 8'.
|
{
"pile_set_name": "Pile-CC"
}
|
Auston Matthews was subject to a drug test after his NHL debut. (Getty Images)
Auston Matthews’ NHL debut was one of the most memorable moments in recent Toronto Maple Leafs history. The team’s first-overall pick in the 2016 draft scored four goals in his first NHL game, putting Toronto’s abysmal 2015-16 season firmly in the rearview mirror.
But, part of the rookie’s celebration involved partaking in the NHL’s randomly selected drug testing the very next day, as Craig Custance of The Athletic writes.
Matthews was at practice with his Toronto teammates when they showed up to give him another welcome to the NHL. He was part of the randomly selected no-notice drug testing that happens during the season under the NHL’s drug policy. The scrutiny from drug testers carried right on into his second season. “For some reason, my name got pulled every single time last year in the season,” Matthews said. “That was insane. Every time they showed up: ‘You. Matthews.’ Every time. I don’t know how many times they came to the rink. I think there was only one time where I didn’t get drug tested.”
On the ice, the young star scored 34-goals and 63-points in 2017-18, but off the ice he also scored his fair share of drug tests.
This season, though, Matthews’ performance should be enhanced only by the addition of John Tavares. Combining with the former New York Islanders’ captain, the pair should form one of the most dangerous 1-2 punches at the centre ice position.
More NHL coverage on Yahoo Canada Sports:
|
{
"pile_set_name": "OpenWebText2"
}
|
Q:
How to Select only Date Hour and Minutes from 2017-09-27 15:39:36.000
I have a column in my table (having value- 2017-09-27 15:39:36.000).
I want to select only Date, Hour, Minutes (i.e:- 2017-09-27 15:39) from column.
Can anyone please tell me how to select.?
A:
Use the FORMAT function to format the date:
SELECT FORMAT(CAST('2017-09-27 15:39:36.000' AS DATETIME), 'yyyy-MM-dd HH:mm')
-- 2017-09-27 15:39
Here is a list of available format specifiers.
|
{
"pile_set_name": "StackExchange"
}
|
Your Message:
Monthly Archives: September 2013
There are so many things that I adore about Darcey & Eric’s wedding! Such as Darcey’s ring with the yellow diamonds that match her yellow Tom’s…. Her beautiful dress, every single one of the bouquets, the flowers in her hair, all the GORGEOUS details that tie in so perfectly with her rustic wedding theme….! I […]
Ahhhh! So I just couldn’t resist posting a few pre-sneak peek images from Darcey & Eric’s AWESOME rustic wedding this last Saturday at Modeana!! The full sneak peek will be up in about a week. Aren’t they just the most adorable couple EVER?!
Karin & Brian are such an awesome couple!!!! They even brought their adorable pup, Baxter, to their engagement session which completely made my day! Just look at how handsome the little guy is with his bow tie and all! My fantastic couple is planning a wedding in San Antonio this coming March. Congratulations, Karin & […]
Awww, yeah! The fall season has started off with a BANG! …and by “BANG” I mean “look at how stinkin’ gorgeous these two are!!!” We started our session in the ever-so-popular & always pretty Arts District. Isabel & Grant share an affinity for lakes and water, so we had to swing by one of my […]
|
{
"pile_set_name": "Pile-CC"
}
|
DOWNLOAD MORE PHOTOS AND VIDEOS @ PARTY ALL STAR!
These starlets are being featured for the first time in skimpy outfits and being naughty. You won’t be disappointed with the high quality photos and videos in the members area. Also, get to interact with the models in a private message board.
Wanna see these girls getting extremely naught? Download the XXX zips right here.
|
{
"pile_set_name": "OpenWebText2"
}
|
Djoemala
Ismail Djoemala (also credited Rd Djoemala; Perfected Spelling: Ismail Jumala; 1915/1918 – 10 June 1992) was an Indonesian actor active in the 1940s. He was often cast alongside Roekiah as her romantic interest.
Biography
Djoemala was born in Batavia (now Jakarta, Indonesia), the capital of the Dutch East Indies. Sources disagree on his year of birth; the catalogue Apa Siapa Film Indonesia gives 19 September 1915, while Indonesian film historian Misbach Yusa Biran gives 1918. Although later billed as "Raden" Djoemala, the actor was not of noble descent. He had some schooling, completing his junior high school in 1934.
A tailor by trade, Djoemala had owned the Broadway Store located on Kramat Raya Street. In 1940 he was approached by Tan's Film and cast for their new production, Sorga Ka Toedjoe (Seventh Heaven). He was cast opposite Tan's mainstay Roekiah, whose regular on-screen partner, Rd Mochtar, had recently left the company owing to wage concerns. Djoemala was chosen for his good looks and tall body. In the film, Djoemala played Hoesin, who reunites the long-separated Hadidjah (Annie Landouw) and Kasimin (Kartolo; Roekiah's husband) to win the hand of Roekiah's character Rasmina. The film was a success, and one reviewer opined that Djoemala was as good as, if not better, than Mochtar.
The pair's next film together, Roekihati, cast Djoemala as a city-dweller named Mansoer who falls for a village girl named Roekihati (Roekiah) but is nearly forced to marry another woman. In 1941 Djoemala and Roekiah acted in another two films together, Poesaka Terpendam (Buried Treasure) and Koeda Sembrani (The Enchanted Horse). Although all of these films were moderate critical successes, ultimately the pairing of Djoemala and Roekiah was unable to draw as many viewers as Mochtar and Roekiah had done. Biran writes that Roekiah's singing and the comedic antics of Kartolo were all that Tan's could depend on during this period.
Following the Japanese occupation of the Indies in February 1942, Tan's was closed. Djoemala is not recorded as having acted in any further films. During the occupation he joined a theatre troupe, Pantjawarna, in West Java, while in the Indonesian National Revolution (1945–49) he fought as a guerrilla. After the revolution he returned to working as an entrepreneur and tailor, dying in Jakarta on 10 June 1992.
Filmography
Sorga Ka Toedjoe (1940)
Roekihati (1940)
Poesaka Terpendam (1941)
Koeda Sembrani (1942/43)
References
Works cited
(book acquired from the collection of Museum Tamansiswa Dewantara Kirti Griya, Yogyakarta)
External links
Category:1910s births
Category:1992 deaths
Category:Year of birth uncertain
Category:20th-century Indonesian male actors
Category:People from Batavia, Dutch East Indies
Category:People from Jakarta
Category:Tailors
|
{
"pile_set_name": "Wikipedia (en)"
}
|
).
2*t**2 + 73
Let r(p) = -2*p**2. Let d(l) = 2141*l + 5. Let q(t) = -3202*t - 7. Let m(w) = 7*d(w) + 5*q(w). What is m(r(i))?
2046*i**2
Let w(l) = -13*l**2 - 54*l - 25. Let g(u) = 2*u - 42. Calculate g(w(n)).
-26*n**2 - 108*n - 92
Let i(b) = -2*b**2 + 12625*b + 30. Let q(f) = -25*f. What is i(q(u))?
-1250*u**2 - 315625*u + 30
Let j(z) = 5*z. Let p(a) = -43607911*a. Calculate p(j(n)).
-218039555*n
Let s(l) = -29*l - 1. Let g be s(-1). Let b(i) = -23 + 76 - 30 - 24. Let o(d) = d + 7. Let w(y) = g*b(y) + 4*o(y). Let q(k) = -4*k**2. What is w(q(c))?
-16*c**2
Let m(b) = -2*b**2. Let o(r) be the third derivative of r**6/24 + 95*r**3/6 + 44*r**2. Let g(z) be the first derivative of o(z). Calculate g(m(u)).
60*u**4
Let g(b) = b. Let f(v) = -2707*v + 9. Let n(r) = -8313196*r + 27642. Let z(m) = 18428*f(m) - 6*n(m). Calculate g(z(q)).
-5420*q
Let g(h) = 11*h. Suppose 9 = 10*n + 3*q, -n + 18 = 5*q + 3. Let f(d) be the first derivative of 0*d + 39 + n*d**2 + 2/3*d**3. Determine f(g(b)).
242*b**2
Let g(d) = -4*d**2 - 2*d**2 + 4*d**2. Let o(z) be the first derivative of 1/2*z**2 - 156 + 0*z. Determine g(o(p)).
-2*p**2
Let j be -3 + (2/(-1) + 109)*-1. Let d = j - -192. Let k(a) = 4*a - 82 + d. Let w(b) = -9*b**2. Calculate w(k(s)).
-144*s**2
Let p(m) = m + 1226 - 215*m - 1224. Let k(d) = 4*d. What is p(k(y))?
-856*y + 2
Let m(t) = 6*t. Suppose -y + 25*y - 240 = 0. Let x(q) = 2546 - y*q - 2546. What is m(x(n))?
-60*n
Let n(y) = 9*y**2 + 2. Let i(u) = 14*u - 38*u + 16*u + 8*u - 14*u**2. Determine i(n(a)).
-1134*a**4 - 504*a**2 - 56
Let u(k) = -2*k + 15. Let b(j) = -15*j**2 - 13*j + 117. Let z(a) = -7*a**2 - 6*a + 54. Let v(g) = -6*b(g) + 13*z(g). What is v(u(n))?
-4*n**2 + 60*n - 225
Let c(f) = 99*f**2. Let j(p) = -15*p - 14. Let n(t) = -6*t - 3. Let q(m) = -j(m) + 5*n(m). Calculate c(q(g)).
22275*g**2 + 2970*g + 99
Let k(a) = 26*a + 2. Let y(v) be the first derivative of -115 - 1/3*v**3 + 0*v + 0*v**2. Give k(y(w)).
-26*w**2 + 2
Let k(a) = a**2. Let w(p) = 16 - 16 + 23*p**2 - 60*p**2 - 44*p**2. Calculate k(w(g)).
6561*g**4
Let h(r) = -22*r**2 + 3. Let q(b) = 4*b**2 + 62290*b - 62290*b. Give h(q(a)).
-352*a**4 + 3
Let c(j) = -5*j**2. Let i(n) = -2*n + 483644. Give i(c(x)).
10*x**2 + 483644
Let b(u) = 16*u + 21*u - 50*u + 11*u. Let d(j) = 10 - 13 - 1 - 20*j**2 + 5. What is b(d(c))?
40*c**2 - 2
Let c(v) = 12*v - 15. Let i(d) = -35*d + 42. Let h(x) = 14*c(x) + 5*i(x). Let u(k) = 55*k. Calculate h(u(t)).
-385*t
Let t(n) = -176*n**2. Let d(m) be the first derivative of -20*m**2 + 5570. Give d(t(k)).
7040*k**2
Let x(b) = 54*b**2. Let q(f) = -1606765*f**2. Give x(q(a)).
139411463322150*a**4
Let k(z) = -41*z**2 - 6*z - 534. Let l(p) = -166*p. Determine k(l(v)).
-1129796*v**2 + 996*v - 534
Let j(n) = -11*n. Let c(i) = i - 30943. Determine j(c(b)).
-11*b + 340373
Let l(z) = 12*z. Let t(m) = 9*m + 20. Let f(q) be the second derivative of 5*q**3/6 + 6*q**2 + 80*q. Let s(d) = 5*f(d) - 3*t(d). Give s(l(h)).
-24*h
Let c(p) = -985*p**2 + 2. Let v(n) = -7*n**2 + 16*n - 8. Let b(h) = 64*h**2 - 152*h + 76. Let w(y) = -2*b(y) - 19*v(y). What is c(w(z))?
-24625*z**4 + 2
Suppose 0 = 3*h + 11 + 4. Let p(w) = w - 1. Let a(o) = 6*o - 7. Let c(f) = h*a(f) + 35*p(f). Let n(l) be the first derivative of -3*l**2/2 + 9. What is n(c(z))?
-15*z
Let c(d) = 8*d + 102. Let o be c(-13). Let j(g) = 8*g - 4. Let h(u) = -u + 2. Let w(z) = o*h(z) - j(z). Let l(v) = -v + 13 - 13. Give l(w(i)).
6*i
Let x(j) = -1307*j. Let l(u) = 24794*u. Give x(l(o)).
-32405758*o
Let f(s) = -5470205*s. Let x(b) = -2*b. Calculate x(f(v)).
10940410*v
Let b be 6/(-14) - (-17)/7. Let g(x) = -2*x**2 - 2*x**2 + x**b - 6*x**2. Let u(y) = 13*y. Calculate u(g(d)).
-117*d**2
Let x(m) = 623*m. Let f(l) = -18*l**2 - 15*l + 7. Calculate x(f(s)).
-11214*s**2 - 9345*s + 4361
Let x(r) = 3*r**2 - 5*r**2 + 0*r**2. Let z(d) = -22*d**2. Let h(b) = 79*b - 248*b + 88*b - 154*b**2 + 81*b. Let n(m) = 3*h(m) - 22*z(m). Give n(x(j)).
88*j**4
Let q(l) be the second derivative of l**4/3 + 8318*l - 2. Let d(n) = 2538*n**2. Determine d(q(t)).
40608*t**4
Let x(j) = 116*j**2. Let v(h) be the second derivative of h**5/120 - 13*h**3/6 - h**2 + 14*h + 2. Let c(a) be the second derivative of v(a). Give c(x(w)).
116*w**2
Let i(a) = 2*a**2. Let t(s) = 173 + 176 + 191 - 540 + 3954*s**2. Give t(i(w)).
15816*w**4
Let u(x) be the third derivative of 0*x**3 + 47*x + 0 + 1/8*x**4 + 2*x**2. Let w(k) = 10*k**2 + 2*k. Give u(w(v)).
30*v**2 + 6*v
Let n(j) = -20*j. Let m(f) = -f. Let q(g) = -4*m(g) - n(g). Let l(w) be the second derivative of 0 + 1/6*w**3 + 0*w**2 - 15*w. What is q(l(p))?
24*p
Let q(u) = -4*u**2. Let s be (-1)/(8/252)*2 + 1. Let m = -60 - s. Let a(r) = -30*r**2 + 0*r + 0*r + 18*r**m. What is a(q(o))?
-192*o**4
Let j(r) = -2*r + 12 + 10 - 21. Let p(d) = 35*d + 15. Let f(z) = -15*j(z) + p(z). Let o(l) = 2*l**2. Calculate o(f(x)).
8450*x**2
Let l(x) = -11*x. Let f(z) = 13*z - 42. Suppose -41*p = -36*p - 75. Let s be f(p). Let n(w) = -s*w + 293*w - 145*w. Calculate l(n(j)).
55*j
Let g(h) = 0*h**2 + 16*h**2 + 10*h**2 + 4*h**2 - 35*h**2. Let o(v) = -26*v - 3. Determine g(o(p)).
-3380*p**2 - 780*p - 45
Let p(i) be the third derivative of i**5/60 + 3411*i**2. Let f(g) = 109*g**2 + 2. Calculate f(p(j)).
109*j**4 + 2
Let t(g) = 55*g - 61*g + 21*g. Let d(z) be the first derivative of z**3 + 17. Determine t(d(r)).
45*r**2
Let n(p) = 2*p**2. Let l(h) be the third derivative of 10363*h**4/24 - 104*h**2 - 24*h. What is n(l(o))?
214783538*o**2
Let t(l) = -1677408*l. Let g(i) = -47*i. Give g(t(a)).
78838176*a
Let o(q) = 2*q**2 + 262*q - 104. Let s(x) = 6*x. Give o(s(y)).
72*y**2 + 1572*y - 104
Let u(r) = 573*r. Let w(g) = 17154*g**2. Give u(w(b)).
9829242*b**2
Let w(d) = -2*d + 9956. Let i(p) = -115*p**2. What is w(i(n))?
230*n**2 + 9956
Let c(o) = 642*o**2. Let b(x) = -473109*x. Calculate c(b(m)).
143700224815602*m**2
Let c(d) = 384*d - 1 + 1 - 196*d - 190*d. Let f(u) = -2. Let x(m) = -14*m - 28. Let i(r) = 70*f(r) - 5*x(r). Determine i(c(p)).
-140*p
Let m(o) = -2*o. Let k = 10473 + -10382. Let n(f) = -988*f. Let a be (2 - -1 - 1)*-1. Let l(h) = -22*h. Let g(w) = a*n(w) + k*l(w). Give m(g(r)).
52*r
Let i(k) = 50*k - 9. Let t(j) = -266*j + 48. Let h(v) = -16*i(v) - 3*t(v). Let x(y) = -686*y. Calculate h(x(u)).
1372*u
Let i(p) = p**2. Let k(z) = -8*z + 85*z + 273*z - 33*z + 53*z. Determine i(k(a)).
136900*a**2
Let j(r) = 8*r - 10. Let v(y) = 26*y - 32. Let l(p) = 16*j(p) - 5*v(p). Let h(n) = -17*n**2 + 12. Determine h(l(c)).
-68*c**2 + 12
Let q(j) = -308*j**2. Let v(x) = 11*x - 17. Let b(m) = 2*m - 3. Let z be -1 + (-5)/((-5)/7). Let h(s) = z*v(s) - 34*b(s). Calculate h(q(d)).
616*d**2
Let v(f) = 2*f**2. Let y be (4 + -5)*3 - -8. Suppose 0 = -4*h + 2*h + 5*d + 26, 3*h - y = -d. Let l(j) = h*j - 57*j + 11*j. Determine l(v(k)).
-86*k**2
Let a(z) = 11*z - 6. Let x(f) = 415*f - 225. Let h(w) = -75*a(w) + 2*x(w). Let p(g) = 643*g**2. Calculate h(p(b)).
3215*b**2
Let c be -2*((-3 - -8) + -7). Let p be c - (-45)/(-10) - 1/(-2). Let f(m) = m + p*m + m. Let a(h) = 7*h. Determine f(a(v)).
14*v
Let m(y) = 8*y**2. Suppose 220*z - 9*z = 5064. Let v(f) be the third derivative of 0*f**3 + 5/24*f**4 + 0*f - z*f**2 + 0. Give v(m(u)).
40*u**2
Let j(o) = o. Let l(v) = 33*v. Let a(c) = 18*j(c) - l(c). Let q(u) = 19341*u - 9663*u - 9679*u. Give q(a(f)).
15*f
Let d(f) = 377*f. Let q(m) = 9692*m**2. Give d(q(c)).
3653884*c**2
Let c(i) = 0*i**2 + 11*i**2 - 5*i**2 - 15*i**2 + 28*i**2. Let v(h) = 3*h**2 + 18*h - 18. Let m(p) = -p + 1. Let x(r) = -18*m(r) - v(r). Give x(c(u)).
-1083*u**4
Let f = -135 - -137. Let w(r) = -f*r + 10227 - 10227. Let c(b) = 32*b + 1. What is c(w(n))?
-64*n + 1
Let v(b) = 10*b. Let x(p) be the second derivative of -p**5/15 - 5*p**3/6 - 9*p**2/2 - 14*p - 1. Let n(s) be the second derivative of x(s). What is v(n(h))?
-80*h
Suppose -4*u = 12 - 36. Let o(t) = 16*t**2 - 4*t**2 - u*t**2 - 10*t**2. Let d(h) = -5*h**2. Give o(d(i)).
-100*i**4
Let g(w) = -49*w - 50*w - 50*w - 33*w + 44 + 184*w. Let y(z) = -48*z. Give y(g(p)).
-96*p - 2112
Let o(a) = 4*a. Let l(n) = 1189*n**2 + 215*n - 19. Determine o(l(d)).
4756*d**2 + 860*d - 76
Let u(z) = z. Suppose -g - 13 = -18. Let d(r) = 2*r**2 - 9*r + 16. Let w be d(g). Let v(b) = w*b - 2 - 8 + 7 + 7. What is v(u(i))?
21*i + 4
Le
|
{
"pile_set_name": "DM Mathematics"
}
|
Dynamin 2 (DNM2) is a mechanochemical and regulatory GTPase that defines the whole class of dynamin-dependent endocytosis mechanisms, including clathrin-mediated (CME), caveolin-mediated, and raft-dependent endocytosis ([@B1][@B2]--[@B3]). It also plays a crucial role in actin and membrane remodelling ([@B4]). Autosomal-dominant centronuclear myopathy (CNM) ([@B5]), dominant Charcot-Marie-Tooth disease (CMT) ([@B6]), and autosomal-dominant hereditary spastic paraplegia (HSP) ([@B7]) are 3 distinct neuromuscular disorders associated with mutations in *DNM2* gene. Also, mutations in *DNM2* were found to be associated with acute lymphoblastic leukemia ([@B8]). *DNM2*-related CNM is a slowly progressive congenital myopathy resulting in generalized muscle weakness with variable severity, ranging from severe neonatal to mild late-onset forms ([@B9][@B10]--[@B11]). CMT is an inherited peripheral neuropathy characterized by muscle weakness and atrophy and loss of touch sensation, which occurs in 1 in 2500 people, making it currently one of the most common incurable inherited neurologic disorders. Mutations in *DNM2* have been identified in the axonal CMT (CMT2) and dominant intermediate forms of CMT ([@B6], [@B12]). HSP is a large group of clinically and genetically heterogeneous disorders resulting in progressive gait disorder because of the dysfunction of long axons in the spinal cord ([@B13]).
Currently, the commonly accepted model of dynamin action in CME depicts a highly orchestrated involvement of all of its domains in the formation and constricting dynamin helix around the neck of invaginated clathrin-coated pits (CCPs) to achieve complete pit separation from the cell membrane and form an endocytic vesicle ([@B14]). Based on the spatial-temporal correlation of clathrin, dynamin, and membrane cargo fluorescence puncta in live cells, it was suggested that dynamin plays a regulatory role in CME, starting from the very early stages of CCP nucleation ([@B15][@B16][@B17]--[@B18]). DNM2 is formed of a GTPase domain that binds and hydrolyses GTP, a middle domain (MD) that is responsible for the assembly of rings and helixes, a pleckstrin-homology (PH) domain that binds phosphoinositides and is responsible for the targeting of dynamin to the plasma membrane, a GTPase effector domain that regulates GTPase activity and self-assembly, and a proline-rich domain mediating protein-protein interactions. To date, 10 heterozygous mutations associated with CMT, 24 with CNM, and 1 in HSP were identified in MD, PH, GTPase effector domain, and proline-rich domain with no common mutations to these 3 disorders.
Based on the early findings that transferrin (Tfn) receptor is internalized predominantly *via* CCPs ([@B19], [@B20]), researchers traditionally used fluorescently labeled or biotinylated Tfn uptake as a measure of CME efficiency. In these assays, the total amount of internalized Tfn molecules is measured in cell lysates or chemically fixed preparations. Several attempts to test whether the disease-associated mutations in *DNM2* affect CME levels using Tfn uptake produced contradictory results. For example, p.R465W mutation in MD has been reported to result in significant reduction in Tfn uptake by Bitoun *et al.*([@B11]) and Koutsopoulos *et al.* ([@B21]), but not by Liu *et al.* ([@B22]) or Sidiropoulos *et al.*([@B23]), who reported no difference in the uptake levels.
The effect of *DNM2* mutations on endocytosis can also be studied by imaging endocytic pits and vesicles in fixed preparations using various electron microscopy (EM) techniques or in living cells by fluorescence live imaging such as confocal or total internal reflection fluorescence microscopy. Though EM is capable of ultra-high resolution sufficient to depict CCP structure at close to molecular level, it cannot provide important information on the kinetics of endocytic vesicle formation. In contrast, fluorescence imaging techniques have an acquisition rate appropriate to follow CCP dynamics in real time but lack the resolution necessary to resolve CCP structure. Advances in super resolution optical imaging, such as stimulated emission depletion (STED), stochastic optical reconstruction microscopy, structured illumination microscopy, and lattice light sheet microscopy (LLSM), now allow for the detection of fluorescently labeled molecules with a resolution of tens of nanometers ([@B24][@B25][@B26]--[@B27]). However, the ability of detecting nothing but fluorescent tags attached to structures of interest with certain affinity is the greatest limitation of fluorescence microscopy. Super resolution fluorescence microscopy has been successfully applied to image endocytosis in live cells ([@B26], [@B28]), although until recently it was difficult to correlate molecular-specific fluorescence data to cell membrane nanostructures. This is because the visualization of the cell membrane topography by live imaging techniques remains challenging. Collectively, the above limitations resulted in remarkable heterogeneity of putative CCPs because fluorescence signals from clathrin contained in genuine CCPs, flat lattices, and endosomes are difficult to distinguish ([@B29], [@B30]). Recently, developments in fluorescence probes for phospholipid components of cell membranes have enabled detailed characterization of vesicle fusion and fission kinetics and helped to elucidate the role of dynamin in fusion pore dynamics in live cells at ∼60 nm spatial and subsecond temporal resolution using STED ([@B31], [@B32]). Also, LLSM was used to follow membrane remodelling and track CCPs and dynamin in live cells at 5.7 s per frame temporal and 150 × 280 nm *xz* resolution ([@B33], [@B34]).
Correlative superresolution fluorescence and metal-replica transmission EM (TEM) has recently been proven suitable for locating fluorescently labeled proteins on the landscape of the cellular plasma membrane at the scale of 20 nm ([@B35]). Although researchers have demonstrated clear colocalization between AF647-labeled clathrin and CCP structures, the technique is not capable of studying CCP kinetics. The lack of live and dynamic investigations in cell membrane morphologic changes, resulting from the recruitment of specific molecules, is a limitation of current imaging techniques, which has been previously highlighted ([@B36], [@B37]).
Previously, we have developed an imaging technique that, in contrast to microscopies described above, enables us to follow morphologic changes of endocytic pits through their entire lifecycle in living cells at close to few nanometers resolution ([@B38]). This technique is based on the combination of scanning ion conductance microscopy (SICM) invented by Hansma and colleagues ([@B39]) and later adapted to be capable of live imaging ([@B40], [@B41]). SICM is a scanning probe microscopy that generates high-resolution topographical images of samples immersed in liquid by raster scanning a glass nanopipette (probe) that follows the surface of a sample very closely without touching. Importantly, SICM imaging does not require any kind of chemical processing of the studied sample and effectively operates in physiologic conditions. This fully noninvasive aspect of imaging makes SICM ideal for living cell studies. The SICM instrument ([**Fig. 1*A***](#F1){ref-type="fig"}) consists of a glass nanopipette mounted on a Z piezo actuator connected to a scan control unit (data not shown) that performs vertical measurements. The same unit also controls raster scanning by moving XY piezo actuator that carries the cell sample in a Petri dish. The ion current that flows between the probing and the reference electrodes located inside the SICM pipette and in the bath, respectively ([Fig. 1*B*](#F1){ref-type="fig"}), is constantly measured by the control electronics. When the pipette vertical position ([Fig. 1*C*](#F1){ref-type="fig"}, top trace) approaches the sample surface to a distance equal to ∼1 radius of pipette opening, the ion current (*I*~pip~) drops ([Fig. 1*C*](#F1){ref-type="fig"}, bottom trace). This drop is used by the control software to stop the approach and record the vertical position of Z piezo actuator as the sample height at a given imaging point. Then, the pipette is withdrawn, and the sample is moved to the next location or pixel. Simultaneous fluorescence confocal imaging is performed by focusing a laser beam at the pipette tip ([@B42], [@B43]). This provides colocalized fluorescence excitation, which is then recorded by a photomultiplier. Confocal autofocusing is achieved by vertical movement of the objective-focusing piezo actuator that is synchronized with the SICM height measurement by Z piezo actuator. This eliminates the need for numerous confocal slicing and subsequent 3-dimensional (3D) reconstruction and allows the fluorescence data to be acquired precisely from the apical membrane in a single scan (red dashed line, [Fig. 1*B*](#F1){ref-type="fig"}). SICM topographical 3D images can be shown either in 2-dimensional view with the height color coded so that the lower areas appear darker and higher areas lighter or as 3D projections. [Figure 1*D*](#F1){ref-type="fig"} shows the example of a 3D topography scan of a cell cytoplasmic membrane where the CCP can be seen as dark invagination (left panel) as well as corresponding fluorescence of clathrin light chain (CLC) a--green fluorescent protein (GFP) (right panel).
{#F1}
Integration of SICM with fluorescence confocal microscopy (FCM) together with 2 orders of magnitude improvement in topographical resolution ([@B38]) and higher imaging rates made it possible to correlate live observations of nanoscale changes in cell membrane morphology with the localized recruitment of fluorescently labeled molecules of interest. Using correlative SICM-FCM, the recruitment of DNM2-GFP to topographically resolved CCPs has been demonstrated for the first time ([@B44]). This technique has also enabled the discovery and characterization of an alternative mechanism of actin-facilitated CCP closure that was more recently confirmed by high-speed atomic force microscopy combined with confocal laser scanning unit ([@B36]). Although temporal resolution of SICM remains slower compared with high-speed atomic force microscopy, STED, and LLSM, its topographical (*i.e.*, spatial) resolution of live cell membranes is higher ([@B45]).
Analysis of the dynamics of nanoscale morphologic changes taking place in a single trafficking event should be critical to decipher the real influence of pathogenic *DNM2* mutations on CME. With this objective, we report here the use of innovative correlative SICM-FCM to study CME in skin fibroblasts from CNM patients harboring 2 distinct heterozygous *DNM2* mutations and in Cos-7 cells expressing corresponding DNM2-GFP mutants.
MATERIALS AND METHODS {#s1}
=====================
SICM instrumentation {#s2}
--------------------
All experiments were performed using the SICM setup that is capable of better than 20 nm topographical resolution on live cells as previously reported ([@B45]). The instrument was controlled by in-house-written software that was also used for data acquisition and analysis. The SICM imaging head controlled by SICM scanner controller (Ionscope, Cambridge, United Kingdom) was built using P-733.2DD XY Piezo-Nanopositioning Stage (PI, Karlsruhe, Germany) 40 µm travel (*xy* movement of the sample) and P-753.21 piezo actuator (PI) 25 µm travel (*z* movement of pipette). Piezo stages were driven by 200 W peak-power low-voltage PZT amplifier E-505 (PI) in capacitive sensor-controlled closed-loop using Sensor & Position Servo-Control Module E-509 (PI). The XY piezo scanner was incorporated into a heavy stainless-steel platform, which was placed onto an inverted Nikon TE2000-U microscope (Nikon, Tokyo, Japan) table spring preloaded and equipped with differential micrometers (OptoSigma, Santa Ana, CA, USA) for precise positioning. Coarse positioning of the pipette in the *z* axis was provided by a M-112.1DG translation stage with a travel range of 25 mm that was coupled with a crossed roller linear translation stage (OptoSigma) to improve stability. To reduce the vibrations caused by the resonance of the glass pipette, standard ESW-F10P holder (Warner Instruments, Hamden, CT, USA) was replaced with a V-grove mounting plate where the pipette was held by a steel spring ∼7 mm above the taper. Nanopipettes were pulled from borosilicate glass (outer diameter, 1 mm; inner diameter, 0.5 mm; Intracell, Saint Ives, United Kingdom) using a laser-based puller Model P-2000 (Sutter Instruments, Novato, CA, USA). The pipettes displayed resistances of ∼400 MΩ (range 300--500 MΩ.) and had an estimated inner diameter of 30 nm. The pipette inner diameters are estimated from the pipette resistance using a half cone angle of 3°. Pipette current was detected *via* an Axopatch 200B (Molecular Devices, Sunnyvale, CA, USA) using a gain of 1 mV/pA and a low-pass filter setting of 5 kHz. The internal holding voltage source of the Axopatch 200B was used to supply a DC voltage of +200 mV to the pipette. The outputs of the capacitive sensors from all 3 piezo elements were monitored using Axon Digidata 1322A (Molecular Devices) and Clampex 9.2 (Molecular Devices). Correlative fluorescence images were recorded using D-104 Microscope Photometer (Photon Technology International, Birmingham, NJ, USA) through a ×100/1.3 oil immersion objective. The excitation was provided by Protera 488 nm wavelength diode-pumped solid-state laser (Laser 200, Huntingdon, United Kingdom). SICM-FCM imaging was done either in Leibovetz's L 15 or CO~2~ Independent medium (Thermo Fisher Scientific, Waltham, MA, USA). Combined fluorescence and topographical imaging was done in Phenol-Red--free Leibovetz's L 15 medium (Thermo Fisher Scientific). The temperature of the medium in the bath was measured with a CL-100 temperature controller (Warner Instruments) as 28 ± 1°C. This temperature that was higher than ambient room temperature was caused by the heat emitted by the piezo scanner, epi-fluorescent attachment, and other surrounding electronic equipment.
Cell culture and plasmids {#s3}
-------------------------
Skin fibroblasts from CNM patients harboring the p.R465W (c.1393C \> T; skin biopsy at the age of 30 yr) or the p.R522H mutation (c.1565G \> A; skin biopsy at the age of 20 yr) and skin fibroblasts from healthy subjects (biopsied at the ages of 26 and 30 for controls 1 and 2) were cultured in DMEM supplemented with 10% fetal calf serum (FCS) in a 5% CO~2~ incubator at 37°C. The p.R465W fibroblast cell line was obtained from a patient of a previously reported large autosomal-dominant CNM family (11), and the p.R522H fibroblast cell line was obtained from a patient of an unpublished autosomal-dominant CNM family. In both patients, mutations were identified by Sanger sequencing.
Monkey Cos-7 cells were routinely maintained at 37°C in 5% CO~2~ using DMEM (Thermo Fisher Scientific) containing 5% (vol/vol) FCS. The plasmid DNA used in the experiments were pCi (Promega, Madison, WI, USA) containing clathrin-GFP (kindly provided by Dr. Lois E. Greene, Laboratory of Cell Biology, National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, MD, USA) ([@B46]). The open reading frame of the wild-type (WT) DNM2 isoform 1 (NM_001005360) was generated by RT-PCR from lymphocyte mRNA and inserted in the frame with the GFP in pcDNA3.1/NT-GFP-TOPO (Thermo Fisher Scientific). The R465W and R522H plasmids were generated by directed mutagenesis using the Quick Change Site-Directed Mutagenesis Kit (Agilent Technologies, Santa Clara, CA, USA). The constructed plasmids were verified by DNA sequencing.
Transfection {#s4}
------------
Cos-7 cells (1 × 10^6^ cells per flask) were plated into a T25 flask and incubated overnight at 37°C in DMEM containing 5% FCS. Cells were washed prior to transfection with PBS, and complexes of Lipofectamine 2000 (Thermo Fisher Scientific) and plasmid DNA at a ratio of 1 μl to 1 μg were added in Opti-MEM (Thermo Fisher Scientific) without FCS to the cells. After 2 h, the medium was replaced with DMEM containing 5% FCS. Cells where then either used for live imaging or fixed for 20 min with 3% formaldehyde containing 5% sucrose.
Uptake assay {#s5}
------------
Human skin fibroblasts and Cos-7 cells were grown on borosilicate 13-mm glass coverslips (VWR International, West Chester, PA, USA) to 60--80% confluency in DMEM and 10% FCS. Tfn from human serum AlexaFluor 647 conjugate (Thermo Fisher Scientific) and CTxB FITC (MilliporeSigma, Burlington, MA, USA) were added to the cells at a 20-µg/ml concentration for 5, 10, 20, and 30 min at 37°C. Cells were washed 3 times with PBS, acid-stripped (0.2 M Na~2~HPO~4~, 0.1 M citric acid), and fixed with 4% paraformaldehyde for 10 min. Coverslips were mounted onto glass slides with prolong diamond antifade mountant with DAPI (Thermo Fisher Scientific). Cells were imaged with Zeiss (Oberkochen, Germany) LSM780 inverted confocal microscope. Individual cells were manually outlined, and corrected total cell fluorescence = integrated density − (area of selected cell × mean fluorescence of background readings) was calculated using plugin contributed by Dr. Martin Fitzpatrick (University of Birmingham, Birmingham, United Kingdom) in Fiji (*<https://fiji.sc/>*) software ([@B47]).
Ultrastructural analysis by TEM {#s6}
-------------------------------
The cells, grown to 80% confluence, were washed twice with PBS and then primary fixed with 2.5% EM-grade glutaraldehyde (Taab Laboratory Equipment, Reading, United Kingdom) in 0.05 M sodium cacodylate buffer (pH 7.2) for 10 min at room temperature. Fixed cells were scraped with a plastic cell scraper and centrifuged at 2000 rpm for 1 min for collection. Supernatant containing glutaraldehyde was removed, and molten agarose (2% wt/vol in distilled water) at 80°C was added to resuspend the cell pellet and left to solidify. The agarose-suspended cell pellets were stored in 2.5% glutaraldehyde in cacodylate buffer at 4°C. The agarose cell pellets were rinsed 3 times in cacodylate buffer for 10 min with gentle rotation. Each agarose pellet was subdivided and postfixed in 1% osmium tetroxide in cacodylate buffer for 1 h. The samples were then washed twice for 5 min in distilled water. Samples were dehydrated by a graded (70--100%) methanol series. The transition to araldite was through propylene oxide. Samples were immersed in 100% propylene oxide twice for 20 min followed by a 50:50 propylene oxide and araldite mixture for 20 min, a 25:75 propylene oxide and araldite mix for 30 min, and 100% araldite for 30 min, and then they were allowed to infiltrate overnight. Samples were then embedded in araldite in molds and polymerization achieved in an embedding oven at 60°C over 72 h. Araldite blocks were trimmed to reveal the embedded sample and 1-µm semithin toluidine blue survey sections cut by microtome (Reichert Ultracut E; Leica Microsystems, Buffalo Grove, IL, USA) for light microscopy. Areas of interest were selected and shaped for ultramicrotomy, and ultrathin (80-nm) sections were cut. The sections were xylene stretched and picked up on 200 mesh thin bar copper grids for TEM. The sections on grids were contrast stained with 1.5% uranyl acetate for 10 min at room temperature and washed 4 times with methanol, followed by staining with lead citrate at room temperature. Observation was by TEM (Hitachi H7000; Hitachi High-Technologies, Tokyo, Japan) operated at 75 kV at the Royal Brompton Hospital.
TEM on metal replica from unroofed cells {#s7}
----------------------------------------
Adherent plasma membranes from human fibroblasts plated on glass coverslips were disrupted by sonication as previously described ([@B48]). Glutaraldehyde- and paraformaldehyde-fixed cells were further sequentially treated with OsO~4~, tannic acid, and uranyl acetate prior to dehydration and Hexamethyldisilazane drying (MilliporeSigma). Dried samples were then rotary-shadowed with platinum and carbon with a high vacuum sputter coater (Leica Microsystems). Platinum replicas were floated off the glass by angled immersion into hydrofluoric acid, washed several times by floatation on distilled water, and picked up on formvar- and carbon-coated EM grids. The grids were mounted in a eucentric side-entry goniometer stage of a transmission electron microscope operated at 80 kV (model CM120; Philips, Andover, MA, USA), and images were recorded with a Morada digital camera (Olympus, Tokyo, Japan). Images were processed in Adobe Photoshop (Adobe, San Jose, CA, USA) to adjust brightness and contrast and presented in inverted contrast.
Western blot {#s8}
------------
Cell pellets were homogenized in lysis buffer containing 50 mM of Tris-HCl pH 7.5, 150 mM NaCl, 1 mM EDTA, and NP40 1% supplemented with protease inhibitor cocktail 1% (MilliporeSigma). After centrifugation (14,000 *g*, 4°C, 15 min), protein concentration in the supernatant was determined with the BCA Protein Assay Kit (Thermo Fisher Scientific). Twenty micrograms of proteins were mixed with loading buffer (50 mM Tris-HCl, SDS 2%, glycerol 10%, β-mercaptoethanol 1%, and bromophenol blue) and denaturated at 90°C for 5 min. Protein samples were separated on SDS--PAGE 10% and transferred onto PVDF membranes (0.45-µm pore size; Thermo Fisher Scientific) overnight at 100 mA at 4°C. Membranes were blocked for 1 h at room temperature in PBS containing nonfat dry milk 5% and Tween 20 0.1% and then exposed to rabbit polyclonal anti-clathrin heavy chain antibody (Abcam, Cambridge, MA, USA), rabbit polyclonal anti-caveolin 1 (Santa Cruz Biotechnology, Dallas, TX, USA), or rabbit polyclonal anti--glyceraldehyde-3-phosphate dehydrogenase antibody (Santa Cruz Biotechnology) in PBS--Tween 20 0.1% and milk 1% overnight at 4°C. Membranes were rinsed in PBS--Tween 20 0.1% and incubated 1 h with horseradish peroxidase--conjugated secondary antibody (anti-rabbit from Jackson ImmunoResearch, West Grove, PA, USA) in PBS--Tween 20 0.1%. Chemiluminescence was detected using ECL Detection Kit (Merck, Darmstadt, Germany) in a G-Box Imaging System (Ozyme, Paris, France), and signal quantification was performed using ImageJ software (National Institutes of Health, Bethesda, MD, USA).
Statistics {#s9}
----------
CCPs lifetimes (LTs) and cell membrane surface area were compared by independent Student's *t* test. CCP densities, fluorescently labeled Tfn and cholera enterotoxin subunit B (CTxB) uptake levels, and Western blots were compared by Mann-Whitney *U* test.
RESULTS {#s10}
=======
In order to investigate how p.R465W (located in MD) and p.R522H (located in PH domain) mutations affect CCP formation, maturation, and closure kinetics, we used the 2 following distinct models. We first confirmed the recruitment of mutant dynamins to the sites of endocytic pit formation and its effect on pit formation, maturation, and closure kinetics in Cos-7 cells transiently transfected with fluorescently labeled DNM2 constructs (DNM2-WT-GFP, DNM2-R465W-GFP, and DNM2-R522H-GFP). Because both CME and endocytosis *via* caveoli are dynamin dependent, DNM2-GFP fluorescence could correlate with topographically detected indentations corresponding to both types of endocytic pits (*i.e.*, CCPs and caveoli). Based on our previously published SICM measurements, average CCP and caveoli opening diameters equal 118 and 70 nm, respectively ([@B42], [@B44]). We have also demonstrated that 88.7% of endocytic pits in Cos-7 cells are clathrin coated, and only 9.35% are caveoli, with the remaining 2% accounting for topographically detected unknown pits that did not have corresponding Clc-GFP or Cav1-GFP fluorescence ([@B42]). Therefore, we conclude that if DNM2-GFP fluorescence colocalizes with topographically detected pit that is larger than 100 nm in diameter, it correlates with CCP. Because transient transfection may result in high DNM2 expression levels that might not reflect physiologically relevant effects and could be cytotoxic ([@B23]), we also studied the impact of *DNM2* mutations on CME in CNM patient--derived skin fibroblasts that have endogenous levels of mutant and WT DNM2. In order to unambiguously identify CCPs in human skin fibroblasts, cells were transfected with CLC-GFP ([@B46]). Based on the previous publications, we assume that because CLC-a itself cannot induce the formation of new CCPs, for which clathrin heavy chain is required, CLC-GFP only labels endogenous CCPs. It would have been impossible to follow the recruitment of mutant DNM2 to the sites of endocytic vesicle formation in human skin fibroblasts transfected with DNM2-GFP because it would interfere with the endogenous mutant DNM2. In both cell types, we measured LT, diameters, and depth of topographically detected pits.
Effect of p.R465W and p.R522H dynamin mutations on CCP formation and scission in Cos-7 cells {#s11}
--------------------------------------------------------------------------------------------
Correlative SICM-FCM time-lapse images of CCPs in Cos-7 cells transiently transfected with DNM2-WT-GFP ([**Fig. 2*A***](#F2){ref-type="fig"}) revealed similar pit behavior to those in Cos-7 cells transfected with CLC-GFP ([@B44]). Topographical SICM images are presented in 2-dimensional view with height color coded so that the lower areas are darker and the higher areas are lighter ([Fig. 2*A*](#F2){ref-type="fig"}, top row). Therefore, CCP is seen as the appearance and disappearance of a dark spot (red arrows) accompanied by corresponding DNM2-WT-GFP fluorescence ([Fig. 2*A*](#F2){ref-type="fig"}, bottom row, and [Supplemental Movie S1](#SM2){ref-type="supplementary-material"}). The entire CCP life cycle, from nucleation through maturation to closure, had the overall LT of 85.1 ± 25.2 s (*n* = 14, 3 different cells, 2 independent experiments). The top panel in [Fig. 2*B*](#F2){ref-type="fig"} shows the cross section profile of the indentation in the cell membrane corresponding to mature CCP in frame 4 ([Fig. 2*A*](#F2){ref-type="fig"}, top row, white dashed line). The mean opening diameter of mature CCPs measured by SICM as full width at half maximum (in practice, half minimum was used because it corresponds to the depth of pit) was 146 ± 11 nm (*n* = 4). [Figure 2*B*](#F2){ref-type="fig"} (bottom panel) shows DNM2-WT-GFP fluorescence profile at maximum intensity. In order to visualize how CCP formation and scission process develops in time, we plotted the sequence of cross section profiles of CCP topography (top) and corresponding DNM2-WT-GFP fluorescence (bottom) assembled in continuous traces ([Fig. 2*C*](#F2){ref-type="fig"}). The traces show clearly that the topographically resolved formation of CCP was followed by the recruitment of DNM2-WT-GFP fluorescence that reached its maximum at the moment of the pit closure and disappeared abruptly right after ([Fig. 2*C*](#F2){ref-type="fig"}, red arrow). Such kinetics of DNM2 recruitment is in agreement with previously published observations ([@B49]).
{#F2}
In Cos-7 cells transiently transfected with DNM2-R465W-GFP, pits successfully went through nucleation to maturation phase, at which they slowed down and started to widen and flatten until finally dissolved ([Fig. 2*D*](#F2){ref-type="fig"} and [Supplemental Movie S2](#SM3){ref-type="supplementary-material"}). The appearance of DNM2-R465W-GFP fluorescence correlated with the topographically detected pit confirms the recruitment of the mutant dynamin to CCPs at the stage of maturation (red arrows). Because of the recruitment of mutant dynamin to CCPs, the LT of CCPs increased significantly compared with control (1931.6 ± 1219.5 s, *n* = 6, 3 cells, 2 experiments, *P* \< 0.005). Also, the recruitment kinetics of mutant dynamin was different from WT, showing high-intensity fluorescence from the very early moments of CCP nucleation. Graphs in [Fig. 2*E*](#F2){ref-type="fig"} show selected, consecutive cross section profiles of 2 adjacent CCPs corresponding to white dashed lines in [Fig. 2*D*](#F2){ref-type="fig"}. The diameter of both pits increased gradually until pits fused into one ∼2-μm--wide, lightly curved structure. By that moment, DNM2-R465W-GFP fluorescence intensity has stopped. Such behavior resembles rather abortive (*i.e.*, non-productive) endocytic events, with pits gradually turning into flat clathrin lattices (FCLs). FCLs were recently described as long-lived structures, also with sustained association with dynamin ([@B50]). Similar to p.R465W, the p.R522H mutant dynamin was targeted to the sites of CCP formation, although at a later stage ([Fig. 2*F*](#F2){ref-type="fig"}, red arrows, and [Supplemental Movie S3](#SM4){ref-type="supplementary-material"}). The LT of CCPs in Cos-7 cells transfected with DNM2-R522H-GFP was also significantly longer compared with control (LT = 928.3 ± 654.4 s, *n* = 7, 4 cells, 4 experiments, *P* \< 0.005). Although not to the same extent as in p.R465W, CCPs in cells expressing p.R522H mutant also widen and flatten prior to disappearance. In our imaging experiments with Cos-7 cells expressing mutant dynamins, we did not observe CCPs with WT-like LTs and formation and closing kinetics (R465W: *n* = 95, 7 cells; R522H: *n* = 116, 8 cells).
Dynamics of CCP formation and scission in skin fibroblasts from CNM subjects with p.R465W and p.R522H dynamin mutations {#s12}
-----------------------------------------------------------------------------------------------------------------------
We next studied CME in human skin fibroblasts because they express endogenous levels of WT (control) or mutant DNM2. The example of typical CCP nucleation, maturation, and closure in human skin fibroblast from a healthy subject transiently transfected with CLC-GFP is shown in [**Fig. 3*A***](#F3){ref-type="fig"} (red arrows) and [Supplemental Movie S4](#SM5){ref-type="supplementary-material"}. Unlike dynamin fluorescence spots that appeared during CCP maturation and peaked at the moment of CCP closure in Cos-7 cells ([Fig. 2*A*](#F2){ref-type="fig"}), CLC-GFP appeared ∼16 s before the topographically detectable invagination corresponding CCP and accompanied the pit throughout its LT in control fibroblasts. The abrupt disappearance of topographically detected indentation that indicates pit closure was followed by the disappearance of associated CLC-GFP fluorescence. The CCP LT was 132.7 ± 85.6 s (*n* = 43, 8 cells, 7 experiments). [Figure 3*B*](#F3){ref-type="fig"} shows a cross section profile corresponding to the white dashed line in [Fig. 3*A*](#F3){ref-type="fig"}, giving pit diameter of 175 nm measured as full width at half maximum. Analysis of CCPs in human skin fibroblasts from patients with p.R465W mutation revealed larger, 380 nm diameter, endocytic structures with significantly longer LTs compared with control fibroblasts (LT = 316.8 ± 220.1 s, *n* = 19, 6 cells, 6 experiments, *P* \< 0.005). A typical example of topographically resolved CCP and corresponding CLC-GFP fluorescence in fibroblast with p.R465W mutation is presented in [Fig. 3*C*](#F3){ref-type="fig"} and [Supplemental Movie S5](#SM6){ref-type="supplementary-material"}. Unlike in cells from healthy individuals, in which the topographically detected pit disappeared abruptly ([Fig. 3*A*](#F3){ref-type="fig"}), in p.R465W cells, the pit gradually widened and flattened ([Fig. 3*C*](#F3){ref-type="fig"}, red arrows). Cross section profiles corresponding to white dashed lines reveal by how much the pit diameter increased and depth reduced during pit degradation ([Fig. 3*D*](#F3){ref-type="fig"}). Such pit behavior is similar to that observed in Cos-7 cells transfected with DNM2-R465W-GFP ([Fig. 2*D, E*](#F2){ref-type="fig"}). In skin fibroblasts from a CNM patient with p.R522H mutation, pits often formed tightly packed clusters ([Fig. 3*E*](#F3){ref-type="fig"} and [Supplemental Movie S6](#SM7){ref-type="supplementary-material"}). In our experiments, together with 16 individually formed pits, we observed 9 clusters formed by 2--4 CCPs where pits nucleate, mature, and close independently in close vicinity. Although the average CCP LT was significantly longer compared with control (LT = 273.4 ± 192.8 s, *n* = 55, 4 cells, 2 experiments, *P* \< 0.005), the abrupt disappearance of individual CCPs and pits in clusters ([Fig. 3*E*](#F3){ref-type="fig"}, red arrow) suggests successful scission events. [Figure 3*F*](#F3){ref-type="fig"} shows the cross section profiles corresponding to white dashed lines in [Fig. 3*E*](#F3){ref-type="fig"}, where 2 static CCPs with longer-than-control LTs can be seen. In our imaging experiments with human skin fibroblasts expressing mutant dynamins, we did not observe CCPs with WT-like LTs and formation and closing kinetics (p.R465W: *n* = 67, 10 cells; p.R522H: *n* = 33, 11 cells).
{#F3}
Characterization of endocytic structures in cytoplasmic membranes of human skin fibroblasts with p.R465W and p.R522H dynamin mutations by EM {#s13}
--------------------------------------------------------------------------------------------------------------------------------------------
In order to further characterize endocytic structures in cells from healthy individuals and subjects with p.R465W and p.R522H mutations, and also to confirm our correlative SICM-FCM observations, we used TEM imaging as a well-established, high-resolution imaging technique. High-resolution TEM images of human skin fibroblasts were surveyed along the cytoplasmic membrane for endocytic structures among which CCPs and caveoli were identified ([**Fig. 4*A--C***](#F4){ref-type="fig"}). CCPs at various stages of formation and internalization could be recognized by the characteristic clathrin coat that, in thin sections, appears darker and thicker than the cytoplasmic membrane (bold white arrows). Numerous caveoli could also be seen as round or flask-shaped invaginations docked to the cell membrane that are nearly 2-times smaller than CCPs in diameter (black arrows). Identified CCPs were binned according to the stage of their life cycle as follows: nucleating (*i.e.*, flat or lightly curved), mature or U-shaped (*i.e.*, invaginated to the depth of more than 1 radius of opening), Ω-shaped (*i.e.*, those with constricted opening), and fully internalized (*i.e.*, detached from the membrane). The distribution of densities of CCPs at different stages of life cycle is presented in [Fig. 4*D*](#F4){ref-type="fig"}. Within the control group of cells, the number of nucleating CCPs and internalized clathrin-coated vesicles was insignificantly higher than mature and Ω-shaped CCPs (299 images, total 130 pits, total membrane length analyzed 3787 μm). Such distribution indicates that CCP closure and scission happens quicker than maturation and transportation deeper inside the cell. In contrast, in p.R465W fibroblasts, the number of internalized vesicles was significantly lower than the number of flat CCPs (227 images, total 90 pits, total membrane length analyzed 3409 μm, *P* \< 0.05) and the number of internalized vesicles in control fibroblasts (*P* \< 0.005). This correlates well with substantially prolonged CCP formation time and eventual pit flattening and disassembly in p.R465W fibroblasts observed by SICM-FCM. In contrast to cells with p.R465W mutation, the distribution of CCPs in p.R522H fibroblasts resembled the one in control cells with an insignificantly higher number of internalized clathrin-coated vesicles (319 images, total 162 pits, total membrane length analyzed 4271 μm). This indicates that despite significantly longer LTs, pits eventually internalize in p.R522H cells as observed by SICM-FCM. The analysis of TEM images was complicated by the fact that the density of CCPs in human skin fibroblasts is low. For comparison, the total numbers of CCPs in control Cos-7 cells is more than 10-times higher. The densities of morphologically identifiable caveoli in human skin fibroblasts were found at least 1 order of magnitude higher than the densities of CCPs at all stages put together (*i.e.*, nucleating through to internalized) in control and both mutant fibroblasts ([Fig. 4*E*](#F4){ref-type="fig"}). The total numbers of identified caveoli in control, p.R465W, and p.R522H fibroblasts were 1906, 1607, and 1626, respectively. The resolution of TEM micrographs was insufficient to identify lightly curved caveoli and distinguish constricted from fully internalized ones; therefore, we could not undertake analysis similar to that of CCPs in full. Hence, we compared the densities of membrane-bound and putatively internalized caveoli ([Fig. 4*A*](#F4){ref-type="fig"}, black arrow and black arrowhead, respectively). In our analysis, only the intracellular vesicles in the vicinity of a half of a micrometer from the morphologically identifiable membrane-bound caveoli were considered internalized caveoli. We found that the total numbers of caveoli in all 3 skin fibroblast lines were roughly similar. Though in control cells, the density of internalized caveoli was the same as membrane-bound ones (*i.e.*, 0.3 ± 0.38 *vs.* 0.3 ± 0.15 caveoli/μm, respectively) ([Fig. 4*F*](#F4){ref-type="fig"}), in p.R465W fibroblasts, the density of internalized caveoli was significantly lower than the membrane-bound ones (0.23 ± 0.3 *vs.* 0.55 ± 0.26 caveoli/μm, *P* \< 0.05), and in p.R522H cells, it was also lower, albeit insignificantly (0.19 ± 0.26 *vs.* 0.45 ± 0.5 caveoli/μm, *P* \> 0.05).
{#F4}
In order to gain further information about the distribution and morphology of CCPs and caveoli at the plasma membrane, EM on metal replica from unroofed cells was performed in healthy control and patient-derived cells. [**Figure 5*A--D***](#F5){ref-type="fig"} illustrates representative images of the cytoplasmic face of the plasma membrane from control cells. CCPs (white arrowheads) and caveoli (white arrows) were clearly identified by their size and their respective coating. In these control cells, caveoli appeared accumulated in some restricted regions ([Fig. 5*B, C*](#F5){ref-type="fig"}). In p.R465W cells ([Fig. 5*E--I*](#F5){ref-type="fig"}) and p.R522H cells ([Fig. 5*J--L*](#F5){ref-type="fig"}), morphology of CCP and caveoli were similar to the control cells, but regions of caveoli accumulation are larger in p.R465W cells ([Fig. 5*F, G*](#F5){ref-type="fig"}).
![TEM views of the cytoplasmic surface of the plasma membrane from unroofed human skin fibroblasts. Representative images from a healthy individual (*A*--*D*), subject with p.R465W mutation (*E*--*I*), and subject with p.R522H mutation (*J*--*L*) are shown. *A*) Survey view of the cytoplasmic surface of the plasma membrane from unroofed primary human fibroblasts of a healthy individual. *B*) Higher magnification view of caveolae-rich region corresponding to the boxed region in *A*. *C*) Higher magnification view of the caveolae-rich region in *B*. *D*) Higher magnification views of CCPs and caveolae from the cell presented in *C*. *E*) Survey view of the cytoplasmic surface of the plasma membrane from unroofed primary human fibroblasts of a subject with p.R465W mutation. *F*) Higher magnification view corresponding to the boxed region in *E*. *G*) Higher magnification view of the caveolae-rich region in *F*. *H*) Survey view of the cytoplasmic surface of the plasma membrane from unroofed primary human fibroblasts of a subject with R465W mutation. *I*) Higher magnification view of CCPs corresponding to the boxed region in *H*. *J*) Survey view of the cytoplasmic surface of the plasma membrane from unroofed primary human fibroblasts of a subject with p.R522H mutation. *K*) Higher magnification view of caveolae-rich region corresponding to the boxed region in *J*. *L*) Higher magnification views of caveolae and CCPs from the cell pictured in *J*. Arrowheads point at CCPs, and small arrows in *D* point at caveolae. Scale bars, 1 µm (*I*, *K*, *L*); 2 µm (*B*, *F*); 10 µm (*A*, *E*, *H*, *J*); 500 µm \[*C*, *D*, *G*, *I* (inset)\].](fj.201802635Rf5){#F5}
Simultaneous uptake of fluorescently labeled Tfn and cholera toxin b in human skin fibroblasts with p.R465W and p.R522H dynamin mutations {#s14}
-----------------------------------------------------------------------------------------------------------------------------------------
In order to confirm whether the mutations in dynamin affect the efficiency of CME and endocytosis by caveoli, we measured simultaneous uptake of AlexaFluor 647-conjugated Tfn from human serum and FITC-conjugated CTxB. This was done assuming that although endocytosis-specific markers such as Tfn receptor, Simian virus, and cholera toxin could be internalized by other independent pathways ([@B51], [@B52]), some preference toward specific pathways exist. Cells were incubated with Tfn and CTxB mix for 5, 10, 20, and 30 min, treated with citric acid to remove surface-bound Tfn and CTxB, fixed, and imaged by Epi-Fluorescence microscopy. Individual cells were manually outlined, and the average fluorescence intensity normalized to a cell surface area was calculated (for details, see Materials and Methods). All cell types were found inhomogeneous in their preference toward Tfn and CTxB uptake ([**Fig. 6*A--D***](#F6){ref-type="fig"}). At 5 and 10 min of incubation, cells didn't show significant difference in uptake. At 20 and 30 min, the uptake of Tfn and CTxB was found to be significantly lower in human skin fibroblasts with mutations in dynamin ([Fig. 6*D, E*](#F6){ref-type="fig"} correspondingly).
{#F6}
Clathrin and caveolin expression levels in human skin fibroblasts with p.R465W and p.R522H dynamin mutations {#s15}
------------------------------------------------------------------------------------------------------------
It has previously been demonstrated that DNM2 expression is unchanged in individuals with CNM ([@B5]). In order to test whether the difference in the total numbers of observed CCPs and caveoli as well as the amounts of fluorescently labeled Tfn and CTxB uptake is not the result of different clathrin and caveolin expression levels, we performed quantitative Western blotting in human skin fibroblasts ([Supplemental Fig. S1](#SM1){ref-type="supplementary-material"}). We found that clathrin heavy chain was expressed at significantly higher levels in p.R522H cells compared with controls and p.R465W, and caveolin 1 was significantly higher in both mutants. The results correlate with the number of endocytic structures observed and the amount of fluorescently labeled cargo uptake.
Morphologic analysis membrane ruffles in apical membranes of human skin fibroblasts induced by p.R465W and p.R522H mutations in DNM2 {#s16}
------------------------------------------------------------------------------------------------------------------------------------
In our previous publication, we have shown that DNM2-GFP colocalizes with roots of highly dynamic microvilli-like protrusions on the cell membrane ([@B44]). With this in mind, we tested whether p.R465W and p.R522H mutations induce morphologic changes at a whole-cell and submicrometer level by performing topographical SICM imaging of living human skin fibroblasts. SICM images of control skin fibroblasts ([**Fig. 7*A***](#F7){ref-type="fig"}) and cells with p.R465W and p.R522H ([Fig. 7*B, C*](#F7){ref-type="fig"}) mutations showed similar looking elongated cells with no obvious morphologic differences. Higher-resolution images ([Fig. 7](#F7){ref-type="fig"}, middle column) revealed surface projections resembling dorsal ruffles and microvilli. Because SICM image is a true 3D map of the apical cell surface composed of a limited number of points, in which each point represents cell height at given at *x* and *y* coordinate, the cell surface area can be calculated as a sum of areas of triangles formed by 3 adjacent SICM scan points ([@B53]). In order to calculate the surface area accumulated in membrane ruffles and microvilli, we first calculated the surface area of the entire cell in raw SICM images, from which we then subtracted the surface area of the same images subjected to low-pass filtration that removed membrane projections. For illustration, cross section profiles of unprocessed (black) and low-pass filtered (red) control cell is presented in [Fig. 7*A*](#F7){ref-type="fig"} (right column, inset). The distributions of the cell membrane ruffles surface area ([Fig. 7](#F7){ref-type="fig"}, right column) revealed that the control population had a higher number of cells rich in microvilli and ruffles compared with mutants. The median surface area accumulated in membrane ruffles and microvilli was 103.65 μm^2^ for control (*n* = 38), insignificantly lower for p.R465W (72.36 μm^2^, *n* = 35, Mann-Whitney *U* test, *P* = 0.083, 2-tailed), and significantly lower for p.R522H cells (72.17μm^2^, *n* = 39, Mann-Whitney *U* test, *P* = 0.012, 2-tailed). Hence, we conclude that p.R522H mutation may result in the reduction of membrane ruffles density and therefore reduction of nonselective forms of uptake.
{#F7}
DISCUSSION {#s17}
==========
The precise mechanism for CME inhibition mediated by CNM and CMT-related DNM2 mutants is still unresolved. Various research techniques used to study this phenomenon produced different types of data that are difficult to link into a coherent model. Using biochemical approaches, it has been demonstrated that p.R465W mutation results in the formation of abnormally stable DNM2 polymers, suggesting that the misbalance in DNM2 assembly-disassembly ratio may disrupt the vesicle release ([@B54]). It has also been suggested that it is the impaired GTPase activity that has the inhibitory effect on CME. However, p.R465W and p.S619L (PH domain) mutants that have increased basal and stimulated GTPase activity either result in reduction or have no effect on the fluorescently labeled cargo uptake putatively specific to CME ([@B54][@B55]--[@B56]).
Using the SICM-FCM that we developed, we found that although in p.R465W mutation, CCPs nucleate as normal, they do not proceed to scission but slowly widen and flatten until they finally disintegrate. This is in contrast to WT DNM2, in which we observed rapid formation and abrupt disappearance of the topographically resolved CCPs accompanied by transient burst and disappearance of DNM2-GFP fluorescence. The recruitment of mutant dynamin to the site of CCP maturation was confirmed by the presence of spatially correlated DNM2-R465W-GFP fluorescence. When expressed in Cos-7 cells, p.R465W mutation results in more than 20-times longer pit LTs. In fibroblasts from individuals with CNM in which p.R465W dynamin is expressed at endogenous levels, we also observed the same effect of CCP flattening, but the LT of CPPs was only twice longer on average, supporting the hypothesis that abnormally high expression levels of dynamin may have detrimental effects. Flat p.R456W CCPs that we observed are similar to recently described FCLs ([@B50]) in their flat profile, prolonged LT, and stable association with dynamin. However, unlike FCLs that are heterogeneous in their shape, p.R465W CCPs retain their round shape through the entire LT. At the early stage of flattening, p.R465W CCPs also share geometrical similarity with clathrin-coated lattices, which were suggested as being capable of productive endocytosis ([@B29]). Significantly lower numbers of fully internalized CCPs in p.R465W skin fibroblasts found in TEM thin sections support our SICM-FCS observations that CME is disrupted in these cells.
Using correlative SICM-FCS, we found that p.R522H mutation also significantly increases the LT of CCPs when expressed in Cos-7 cells and in patient-derived skin fibroblasts. However, in contrast to p.R465W, p.R522H CCPs maintain their shape and close abruptly, indicating that productive endocytosis takes place. Thin-section TEM data also suggest that CME is not strongly affected by p.R522H mutation, showing the densities of internalized CCPs in p.R522H fibroblasts being slightly higher compared with control. This observation is counterintuitive considering significantly longer LT of CCPs and could only be explained by a concomitant defect in uncoating. In our SICM-FSC observations, we also found unusual clustering of CCPs. Such closely formed CCPs were reported in early studies of fibroblasts by quick-freeze EM ([@B57]). More recent studies on the dynamics of CCPs using total internal reflection fluorescence (TIRF) microscopy were mainly focused on characterization of individual, sparsely placed pits, possibly because of lack of resolving power. Clustering observed in our SICM-FCM experiments and EM images of metal replica from unroofed cells was not observed in Cos-7 cells expressing DNM2-R522H-GFP, suggesting that it is not the mutation *per se* that results in clustering of CCPs.
Our TEM images revealed high densities of membrane-bound and internalized caveoli in control and mutant human skin fibroblasts. The fact that we did not observe numerous indentations corresponding to caveolar openings in SICM topographical images could either be explained by tightly constricted caveolar necks, making the size of the opening below the resolution level of our SICM in its present setting, or the presence of stomatal diaphragms ([Fig. 4*A*](#F4){ref-type="fig"}, right panel, asterisk) formed by PV1 protein encoded by plasmalemmal vesicle--associated protein gene in humans that is known to be expressed in skin ([@B58]). Such diaphragms that form at caveolar openings could physically prevent the SICM nanopipette from entering caveoli. Therefore, the spatial resolution of correlative SICM-FCS requires further improvement in order to make topographical imaging of caveoli possible.
In order to enable better comparison of our findings with previously published observations, we also measured the uptake of fluorescently labeled Tfn and CTxB and found that it is reduced in cells with p.R465W and p.R522H. Although uptake data support our findings at individual pit level and indicate that both CME levels are reduced by these mutations, the results have to be treated with caution because a substantial proportion of Tfn is uptaken by clathrin- and dynamin-independent mechanisms. Indeed, cells transfected with GTPase-deficient K44A mutant dynamin or treated with dynamin inhibitor Dynasore, often used as a negative control, on average show only 70% reduction in uptake at best ([@B11], [@B22]). This is compared with nearly complete absence of internalization on ice when all other remaining pathways are blocked ([@B21]). Such low specificity of fluorescently labeled cargo toward 1 particular uptake mechanism and the existence of alternative internalization pathways is known for currently used endocytic markers (*e.g.*, cholera toxin B) ([@B59]) and Tfn receptor ([@B60], [@B61]). Other nonspecific forms of uptake, such as micropinocytosis facilitated by membrane ruffles, in which dynamin is involved also contribute to the complexity. Indeed, it has previously been demonstrated that dynamin directly interacts with actin, and point mutations in the actin-binding domain cause aberrant membrane ruffling ([@B62]). In particular, it has been shown that p.R465W mutation significantly decreases the enrichment to the dorsal ruffle and supresses raft-dependent endocytosis; however, this does not affect the rate of macropinocytosis of horse radish peroxidase ([@B23]). It is also well documented that DNM2 plays a crucial role in macropinocytosis ([@B63], [@B64]) and acts as a transition controller for the recruitment of actin-related protein-2/3 (ARP2/3) complex activators required for IL-2 receptor endocytosis by membrane protrusion-assisted clathrin- and caveolin-independent mechanisms ([@B65]). Using SICM topographical imaging alone, we have demonstrated that mutations in DNM2 significantly reduce membrane ruffle density and therefore may affect nonselective uptake levels, further complicating the interpretation of uptake of fluorescent markers specific to particular types of endocytosis. This correlates well with the recently published observation that CNM-linked DNM2 mutations disrupt the formation of new actin filaments as well as the stimulus-induced translocation of glucose transporter 4 to the plasma membrane ([@B66]).
Our study suggests that CME and caveolae function are impaired by CNM-associated DNM2 mutations, which may have direct clinical relevance. On one hand, caveolae dysfunction was associated with a class of diseases called caveolinopathies, encompassing a wide range of clinical spectrum ([@B67]), including skeletal muscular diseases ([@B68]). In CNM, abnormal caveolin staining and accumulation of caveolae have already been reported in muscle biopsies from patients affected by the autosomal recessive form because mutations in the bridging integrator 1 (*BIN1*) gene encoding the Amphiphysin 2 ([@B69], [@B70]). We report here the first evidence that CNM-associated DNM2 mutations may also affect caveolae function. Further investigation will be necessary to confirm such a defect in muscle tissue and identify potential consequences on caveolae functions, including endocytosis, plasma membrane organization, signaling pathways, and mechanosensing. On the other hand, alteration of CME, a key regulator of turnover and function of membrane receptors and channels, may also lead to a deleterious impact on muscle function. A defect of clathrin-coated vesicles has been previously associated with several human diseases, including cancer, neurologic disorders, and myopathy ([@B71], [@B72]), and remains to be further investigated in centronuclear myopathies. In particular, CME defect may alter membrane content of ion channel, as suggested by alteration of the membrane permeability to calcium, leading to abnormal calcium homeostasis in a mouse model of the DNM2-linked CNM ([@B73]). A better knowledge of involvement of clathrin- and caveolin-dependent endocytosis in the pathomechanisms of the DNM2-linked CNM will be of particular interest for identification of specific targets for future development of therapeutic approaches.
Overall, in this study, we showed how SICM-FCM is able to establish a spatial and temporal correlation between the formation of the endocytic pit nanostructure and the recruitment of fluorescently tagged molecules responsible for the pit formation and scission. This is a highly important advance that makes dissecting the mechanistic action of membrane proteins possible in real time and at nanoscale. Here, we demonstrated the impairment of CME as potential pathomechanism in *DNM2*-related CNM at the structural level of individual CCPs in the most pertinent cell model (*i.e.*, patient-derived cells).
Supplementary Material {#s19}
======================
This article includes supplemental data. Please visit *<http://www.fasebj.org>* to obtain this information.
######
Click here for additional data file.
######
Click here for additional data file.
######
Click here for additional data file.
######
Click here for additional data file.
######
Click here for additional data file.
######
Click here for additional data file.
######
Click here for additional data file.
This article includes supplemental data. Please visit *<http://www.fasebj.org>* to obtain this information.
The authors thank Dr. P. Novak (Queen Mary University of London) for scanning software development, Dr. Andy Rogers (Royal Brompton and Harefield NHS Trust) for expertise in TEM, and Dr. Benedict Reilly-O'Donnell (National Heart and Lung Institute, Department of Cardiac Medicine, Imperial College London, London, United Kingdom) for proofreading of the manuscript. This work was supported by PhD studentship to T.A. from Muscular Dystrophy UK (Grant P49914 to A.S.), Biotechnology and Biological Sciences Research Council (BBSRC) funding to A.S. (BB/M022080/1), the Agence Nationale de la Recherche (Grant ANR-14-CE12-0009 to M.B. and Young Researcher Grant ANR-14-CE12-0001-01 to S.V.), and British Heart Foundation (BHF) funding for J.G. and I.A.D. (RG/17/13/33173). A.S. and Y.K. are shareholders in ICAPPIC, Ltd., a company commercializing nanopipette-based instrumentation. The authors declare no other conflicts of interest.
3D
: 3 dimensional
CCP
: clathrin-coated pit
CLC
: clathrin light chain
CNM
: centronuclear myopathy
CME
: clathrin-mediated endocytosis
CMT
: Charcot-Marie-Tooth disease
CTxB
: cholera enterotoxin subunit B
DNM2
: dynamin 2
EM
: electron microscopy
FCL
: flat clathrin lattice
FCM
: fluorescence confocal microscopy
FCS
: fetal calf serum
GFP
: green fluorescent protein
HSP
: hereditary spastic paraplegia
LLSM
: lattice light sheet microscopy
LT
: lifetime
MD
: middle domain
PH
: pleckstrin-homology
TEM
: transmission electron microscopy
SICM
: scanning ion conductance microscopy
STED
: stimulated emission depletion
Tfn
: transferrin
WT
: wild type
AUTHOR CONTRIBUTIONS {#s18}
====================
M. Bitoun and A. Shevchuk designed the experiments; T. Ali, J. Bednarska, S. Vassilopoulos, M. Tran, and A. Shevchuk performed experiments; T. Ali, I. A. Diakonov, and A. Ziyadeh-Isleem prepared plasmids and cell culture; T. Ali, S. Vassilopoulos, Y. E. Korchev, M. Bitoun, and A. Shevchuk analyzed data; and T. Ali, P. Guicheney, J. Gorelik, M. M. Reilly, M. Bitoun, and A. Shevchuk wrote the manuscript.
|
{
"pile_set_name": "PubMed Central"
}
|
/*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package org.apache.hadoop.hive.llap.daemon.impl;
import org.apache.hadoop.metrics2.AbstractMetric;
import org.apache.hadoop.metrics2.MetricsCollector;
import org.apache.hadoop.metrics2.MetricsInfo;
import org.apache.hadoop.metrics2.MetricsRecordBuilder;
import org.apache.hadoop.metrics2.MetricsTag;
import java.util.Map;
public final class DumpingMetricsCollector implements MetricsCollector {
private MetricsRecordBuilder mrb;
public DumpingMetricsCollector(Map<String, Long> data) {
mrb = new DumpingMetricsRecordBuilder(data);
}
@Override
public MetricsRecordBuilder addRecord(String s) {
return mrb;
}
@Override
public MetricsRecordBuilder addRecord(MetricsInfo metricsInfo) {
return mrb;
}
private final class DumpingMetricsRecordBuilder extends MetricsRecordBuilder {
private Map<String, Long> data;
private DumpingMetricsRecordBuilder(Map<String, Long> data) {
this.data = data;
}
@Override
public MetricsCollector parent() {
throw new UnsupportedOperationException();
}
@Override
public MetricsRecordBuilder tag(MetricsInfo metricsInfo, String s) {
return this;
}
@Override
public MetricsRecordBuilder add(MetricsTag metricsTag) {
return this;
}
@Override
public MetricsRecordBuilder add(AbstractMetric abstractMetric) {
throw new UnsupportedOperationException();
}
@Override
public MetricsRecordBuilder setContext(String s) {
return this;
}
@Override
public MetricsRecordBuilder addCounter(MetricsInfo metricsInfo, int i) {
data.put(metricsInfo.name(), Long.valueOf(i));
return this;
}
@Override
public MetricsRecordBuilder addCounter(MetricsInfo metricsInfo, long l) {
data.put(metricsInfo.name(), l);
return this;
}
@Override
public MetricsRecordBuilder addGauge(MetricsInfo metricsInfo, int i) {
data.put(metricsInfo.name(), Long.valueOf(i));
return this;
}
@Override
public MetricsRecordBuilder addGauge(MetricsInfo metricsInfo, long l) {
data.put(metricsInfo.name(), l);
return this;
}
@Override
public MetricsRecordBuilder addGauge(MetricsInfo metricsInfo, float v) {
data.put(metricsInfo.name(), Long.valueOf(Math.round(v)));
return this;
}
@Override
public MetricsRecordBuilder addGauge(MetricsInfo metricsInfo, double v) {
data.put(metricsInfo.name(), Math.round(v));
return this;
}
}
}
|
{
"pile_set_name": "Github"
}
|
Q:
Why are output nets also required to be redeclared as either 'wire' or 'reg'?
Why do we have to take the same variable name of an output and also wire for getting the value? eg:
module TEST(INP1,INP2,CIN,COUT,SUM);
input [31:0] INP1;
input [31:0] INP2;
output [31:0] SUM;
input CIN;
output COUT;
wire [31:0] SUM;// Again redefined
wire COUT; // Again Redefined
assign {COUT,SUM} = INP1 + INP2 + CIN ;
Example for getting the Carry-out and the Sum of two numbers and Carry-In taken as the input.
A:
It is not necessary to declare an output also as a wire. Furthermore, you can avoid duplicating the port list by using ANSI-stlye port declarations:
module TEST (
input [31:0] INP1,
input [31:0] INP2,
output [31:0] SUM,
input CIN,
output COUT
);
assign {COUT,SUM} = INP1 + INP2 + CIN ;
endmodule
In your example, you do not need to declare outputs as reg. But, if you need to for another circuit, you can declare the type on the same line, such as:
output reg [31:0] Q;
A:
Verilog 1995 did require the port direction to be listed after. Output wire types were implicit and regs could be declared inline with direction.
module TEST(A,B,C,D);
input [31:0] A;
input [31:0] B;
output [31:0] C;
output D;
reg D;
could be written as:
module TEST(A,B,C,D);
input [31:0] A;
input [31:0] B;
output [31:0] C;
output reg D; //Only declared twice
Since Verilog 2001 the extra definition is no longer required and they can be declared inline (ANSI-Style).
module TEST(
input [31:0] A,
input [31:0] B,
output [31:0] C,
output reg D // Declared Once
);
From SystemVerilog (2009) we have the logic type, you no longer have to switch between reg and wire types. The only requirement is that if you need to tri-state use wire or tri.
module TEST(
input [31:0] A,
input [31:0] B,
output logic [31:0] C,
output logic D
);
My understanding of the original requirement for having reg and wire types was for simulation speed or ease of simulator design. The value of a wire is evaluated every simulation delta cycle while a reg is only evaluated when triggered by the sensitivity list.
|
{
"pile_set_name": "StackExchange"
}
|
So, you want to start building your own coils? It makes vaping more fun! So many more choices in build styles, flavor productions, vapor productions, etc.! Building is my absolute favorite part of vaping! I will be taking you through step-by-step of how to build your basic coil for your RDA/RTA/ RDTA ! (P.S. if you want to watch the video instead of reading this, click here
You will need all the materials necessary for building, You will need your wire, cotton, post to wrap your wire around, scissors, and ceramic tweezers (Shown Below). See our selection of tools and materials here In this picture demonstration, I will be using the tools included in the Coil Master, a wonderful kit if you want to make building as easy as possible!
One you have all your materials, choose the post you want to wrap the coils around. I used a 3mm post in this particular demonstration, since it is the most common.
Insert your wire through the coil jig and place the top on. Spin the top to easily wrap the wiring around the post to make your coil. (make sure you do not apply too much pressure, as this could cause problems while wrapping)
Something to think about when choosing wire is the gauge. The lower gauge wire (22g, 24g, 26g) is lower resistance wire, so you can use more wire without increasing the resistance of the coil. This is useful when making lower ohm builds! The more surface area, the more vapor produced.
The number of wraps will also affect the resistance of the coil! You can either look up the gauge wire and wraps needed for a certain ohm, or just play with the builds to figure it out on your own!
Once the coil is wrapped and ready to go, use the wire cutters to clip off any extra wire. (make sure you leave about an inch or so on each side so you can thread the coil into the deck)
Once you have the coil in place, tighten the wire down. This can be different depending on the style deck. I am using a clamp style deck in this process, but there are multiple different deck types.
After tightening the wire down, use the wire cutters to clip of the excess wire. Once this is done, make sure the coil is centered and straightened for proper firing and wicking.
If you are using a regulated mod , which most new builders would, the mod will read the resistance (ohm) of the coil. This is where you can choose the wattage you want to run the device at. Refer to my Ohms and Wattage article for more information on this subject.
Now that the coil is set into place, you have to fire your device to see if the coil heats up properly. It’s almost impossible to build a perfectly firing coil, so you will need your ceramic tweezers to spread or squeeze the coil until it all heats up evenly. You should also pay attention at this stage to making sure that your coils are STRAIGHT & CENTERED in the deck. If you don't do this it can lead to dry hits because your cotton is not wicking juice evenly from both sides of the coil. This is the trickiest part for beginners, as it can take some practice to get it perfect.
When the coil(s) are heating up correctly, you are now ready for wicking. Take your cotton and fit it into your coil(s). Cut off any unneeded extra and tuck the cotton to the base of the deck. Depending on style, you may need to move cotton to sides or center for best airflow.
If you are making dual coils, repeat all above steps!
|
{
"pile_set_name": "OpenWebText2"
}
|
Trade Deadline Target: Hunter Pence
Receive the latest red-fever updates in your inbox
The Texas Rangers have been serious players at the MLB trade deadline since Jon Daniels became the club's general manager, and this year will be no different. Let's take a look at a possible Rangers target for the July 31 deadline.
Target:Hunter Pence, OF, San Francisco Giants
The Giants have said they will not listen to trade offers for the veteran outfielder, who also happens to be an Arlington native and a graduate of Arlington High School who went on to play at UTA and broke into the league with the Houston Astros.
But that makes absolutely no sense. The Giants are not going to win the NL West and probably won't make the playoffs. Oh, and Pence is one of those good ol' trade-deadline rentals as he'll hit free agency this winter. It makes no sense if the Giants shut down talks on Pence before they even begin, so it's most likely just a leverage move by the Giants' front office to build up a market for Pence, who's a solid player and a great team leader.
Pence is hitting .268 with 14 homers and 50 RBIs this season and is 14-for-14 on stolen-base attempts this season. That would give the Rangers exactly what they need — another bat on the right side with some pop who can play the corner outfield and relieve David Murphy of being an everyday player. Also, if Nelson Cruz gets suspended, the Rangers would have some insurance in right field.
Pence makes perfect sense for the Rangers, and dealing Pence should make perfect sense for the Giants.
|
{
"pile_set_name": "Pile-CC"
}
|
Priorities in the management of polytraumatised patients with head injury: partially resolved problems.
The results of a retrospective analysis of a series of 73 polytraumatised patients with severe head injuries are presented. The neurological outcome has been less favourable in cases with early osteosynthesis of peripheral fractures compared to cases in which such operations had been delayed. Therefore it seems advisable to postpone the non-vital operative treatment of osteoarticular lesions until stabilisation of the cerebral and vegetative situation.
|
{
"pile_set_name": "PubMed Abstracts"
}
|
MADRID — A Spanish judge on Monday ordered the release on bail of six Catalan separatist politicians, just three weeks before regional elections, while keeping four others in custody.
The ruling, which was condemned by Catalan separatists, pointed to the complications of the campaign, which officially opens Tuesday, before the Dec. 21 elections for a new regional Parliament.
Prime Minister Mariano Rajoy of Spain called the elections in the hope that voters would replace the separatists who had used their fragile majority to push through an October referendum on independence that Spanish courts declared illegal.
The central government in Madrid has been administering the region directly since removing the region’s leadership. The former Catalan leader Carles Puigdemont left for Belgium rather than face Spain’s judiciary.
|
{
"pile_set_name": "OpenWebText2"
}
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.