This repository hosts the official implementation of GEARS, a method that can predict transcriptional response to both single and multi-gene perturbations using single-cell RNA-sequencing data from perturbational screens.
Install PyG, and then do pip install cell-gears.
Using the API, you can (1) reproduce the results in our paper and (2) train GEARS on your perturbation dataset using a few lines of code.
from gears import PertData, GEARS
# get data
pert_data = PertData('./data')
# load dataset in paper: norman, adamson, dixit.
pert_data.load(data_name = 'norman')
# specify data split
pert_data.prepare_split(split = 'simulation', seed = 1)
# get dataloader with batch size
pert_data.get_dataloader(batch_size = 32, test_batch_size = 128)
# set up and train a model
gears_model = GEARS(pert_data, device = 'cuda:8')
gears_model.model_initialize(hidden_size = 64)
gears_model.train(epochs = 20)
# save/load model
gears_model.save_model('gears')
gears_model.load_pretrained('gears')
# predict
gears_model.predict([['FOX1A', 'AHR'], ['FEV']])
gears_model.GI_predict([['FOX1A', 'AHR'], ['FEV', 'AHR']])To use your own dataset, create a scanpy adata object with a gene_name column in adata.var, and two columns condition, cell_type in adata.obs. Then run:
pert_data.new_data_process(dataset_name = 'XXX', adata = adata)
# to load the processed data
pert_data.load(data_path = './data/XXX')| Name | Description |
|---|---|
| Dataset Tutorial | Tutorial on how to use the dataset loader and read customized data |
| Model Tutorial | Tutorial on how to train GEARS |
| Plot top 20 DE genes | Tutorial on how to plot the top 20 DE genes |
| Uncertainty | Tutorial on how to train an uncertainty-aware GEARS model |
@article {Roohani2022.07.12.499735,
author = {Roohani, Yusuf and Huang, Kexin and Leskovec, Jure},
title = {GEARS: Predicting transcriptional outcomes of novel multi-gene perturbations},
year = {2022},
doi = {10.1101/2022.07.12.499735},
publisher = {Cold Spring Harbor Laboratory},
journal = {bioRxiv}
}
Preprint: Link
Code for reproducing figures: Link
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent