Update README.md
Browse files
README.md
CHANGED
@@ -5,4 +5,74 @@ task_categories:
|
|
5 |
- text-classification
|
6 |
tags:
|
7 |
- biology
|
8 |
-
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
5 |
- text-classification
|
6 |
tags:
|
7 |
- biology
|
8 |
+
---
|
9 |
+
|
10 |
+
Dataset Summary
|
11 |
+
This repository contains datasets derived from UniProt and SwissProt databases for analyzing biological sequences.
|
12 |
+
The data includes raw sequences, processed files, and benchmark datasets split using hierarchical strategies to ensure diversity and
|
13 |
+
discourage model bias. The datasets are organized into different folders for balanced and unbalanced splits, with additional variations
|
14 |
+
where full-length sequences are chunked into smaller sequences.
|
15 |
+
|
16 |
+
Dataset Structure
|
17 |
+
The file directory is structured as follows:
|
18 |
+
|
19 |
+
UniprotAndSwissprotDatasets:
|
20 |
+
SwissprotDatasets:
|
21 |
+
BalancedSwissprot: train.csv, validation.csv
|
22 |
+
UnbalancedSwissprot: train.csv, validation.csv
|
23 |
+
**test.csv, test1.csv, test2.csv`
|
24 |
+
SwissprotDatasetsChunked:
|
25 |
+
Contains the same data as SwissprotDatasets, but each full-length sequence is split into smaller sequences of 20 read length.
|
26 |
+
BalancedSwissprot: train.csv, validation.csv
|
27 |
+
UnbalancedSwissprot: train.csv, validation.csv
|
28 |
+
**test.csv, test1.csv, test2.csv`
|
29 |
+
UniprotDatasets:
|
30 |
+
BalancedUniprot: train.csv, validation.csv
|
31 |
+
UnbalancedUniprot: train.csv, validation.csv
|
32 |
+
**test.csv, test1.csv, test2.csv`
|
33 |
+
UniprotDatasetsChunked:
|
34 |
+
Contains the same data as UniprotDatasets, but each full-length sequence is split into smaller sequences of 20 read length.
|
35 |
+
BalancedUniprot: train.csv, validation.csv
|
36 |
+
UnbalancedUniprot: train.csv, validation.csv
|
37 |
+
**test.csv, test1.csv, test2.csv`
|
38 |
+
Data Description
|
39 |
+
The datasets are derived from:
|
40 |
+
|
41 |
+
UniProt: Contains two main sections:
|
42 |
+
TrEMBL: Automatically annotated; contains homology-based annotations with more errors.
|
43 |
+
SwissProt: Manually curated with experimental evidence of functional annotations.
|
44 |
+
Preprocessing Steps:
|
45 |
+
Organism Filtering: Only prokaryotic organisms (bacteria and archaea) were included.
|
46 |
+
Data Mapping: UniProt IDs were mapped to UniRef clusters (UniRef50, UniRef90, UniRef100) and EMBL CDS IDs.
|
47 |
+
Data Cleaning:
|
48 |
+
Removed records with missing or partial EC numbers.
|
49 |
+
Split records with multiple EC numbers into individual entries.
|
50 |
+
Dataset Splitting
|
51 |
+
A hierarchical splitting strategy ensured sequences from the same cluster do not appear in more than one set:
|
52 |
+
|
53 |
+
Train, Validation, and Test Sets are derived based on UniRef50, UniRef90, or UniRef100 clusters.
|
54 |
+
Special test sets include:
|
55 |
+
In-Distribution (Test Set-I): Contains sequences with EC numbers present in the training sets.
|
56 |
+
Out-of-Distribution (Test Set-II): Contains sequences with EC numbers absent from the training sets.
|
57 |
+
Benchmarks
|
58 |
+
Four benchmark datasets were created:
|
59 |
+
|
60 |
+
Benchmark-I: SwissProt+TrEMBL (unbalanced).
|
61 |
+
Benchmark-II: SwissProt+TrEMBL (balanced).
|
62 |
+
Benchmark-III: SwissProt only (unbalanced).
|
63 |
+
Benchmark-IV: SwissProt only (balanced).
|
64 |
+
Chunked Sequences
|
65 |
+
ChunkedSwissProt and ChunkedUniProt folders contain the same data as their respective non-chunked counterparts but with each full-length sequence divided into smaller 20 read-length sequences.
|
66 |
+
These chunked datasets are especially useful for sequence alignment and similarity-based tasks.
|
67 |
+
Dataset Generation
|
68 |
+
To generate the dataset, follow the instructions in the notebooks:
|
69 |
+
|
70 |
+
DatasetGeneration.ipynb: Processes raw data to create final datasets.
|
71 |
+
DatasetSplitting.ipynb: Splits the dataset into training, validation, and test sets.
|
72 |
+
Citation
|
73 |
+
If using this dataset, please cite the original UniProt and SwissProt resources:
|
74 |
+
|
75 |
+
UniProt: https://www.uniprot.org/
|
76 |
+
SwissProt: https://www.uniprot.org/uniprotkb?query=reviewed:true
|
77 |
+
License
|
78 |
+
This dataset is licensed under the Apache 2.0 License.
|