Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,29 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
---
|
2 |
dataset_info:
|
3 |
features:
|
|
|
1 |
+
To accompany OpenPhenom, Recursion is releasing the **RxRx3-core** dataset, a challenge dataset in phenomics optimized for the research community.
|
2 |
+
RxRx3-core includes labeled images of 735 genetic knockouts and 1,674 small-molecule perturbations drawn from the [RxRx3 dataset](https://www.rxrx.ai/rxrx3),
|
3 |
+
image embeddings computed with [OpenPhenom](https://huggingface.co/recursionpharma/OpenPhenom), and associations between the included small molecules and genes.
|
4 |
+
|
5 |
+
Mapping the mechanisms by which drugs exert their actions is an important challenge in advancing the use of high-dimensional biological data like phenomics.
|
6 |
+
We are excited to release the first dataset of this scale probing concentration-response along with a benchmark and model to enable the research community to
|
7 |
+
rapidly advance this space.
|
8 |
+
|
9 |
+
Loading the RxRx3-core image dataset
|
10 |
+
```
|
11 |
+
from datasets import load_dataset
|
12 |
+
rxrx3_core = load_dataset("recursionpharma/rxrx3-core")
|
13 |
+
```
|
14 |
+
Loading OpenPhenom embeddings and metadata for RxRx3-core
|
15 |
+
```
|
16 |
+
from huggingface_hub import hf_hub_download
|
17 |
+
import pandas as pd
|
18 |
+
|
19 |
+
file_path_metadata = hf_hub_download("recursionpharma/rxrx3-core", filename="metadata_rxrx3_core.csv",repo_type="dataset")
|
20 |
+
file_path_embs = hf_hub_download("recursionpharma/rxrx3-core", filename="OpenPhenom_rxrx3_core_embeddings.parquet",repo_type="dataset")
|
21 |
+
|
22 |
+
open_phenom_embeddings = pd.read_parquet(file_path_embs)
|
23 |
+
rxrx3_core_metadata = pd.read_csv(file_path_metadata)
|
24 |
+
```
|
25 |
+
Benchmarking code for this dataset is provided in the [EFAAR benchmarking repo](https://github.com/recursionpharma/EFAAR_benchmarking/tree/trunk).
|
26 |
+
|
27 |
---
|
28 |
dataset_info:
|
29 |
features:
|