Awesome

scGen predicts single-cell perturbations.

This repository includes python scripts in code and notebooks in the Jupyter Notebooks folder to reproduce figures from the paper (bioRxiv, 2018) according to the table bellow.

figure	notebook path
Figure 2	Jupyter Notebooks/Fig2.ipynb
Figure 3	Jupyter Notebooks/Fig3.ipynb
Figure 4	Jupyter Notebooks/Fig4.ipynb
Figure 5	Jupyter Notebooks/Fig5.ipynb
Figure 6	Jupyter Notebooks/Fig6.ipynb
Supplemental Figure 1	Jupyter Notebooks/SupplFig1.ipynb
Supplemental Figure 2	Jupyter Notebooks/SupplFig2.ipynb
Supplemental Figure 4	Jupyter Notebooks/SupplFig4.ipynb
Supplemental Figure 5	Jupyter Notebooks/SupplFig5.ipynb
Supplemental Figure 6	Jupyter Notebooks/SupplFig6.ipynb
Supplemental Figure 7	Jupyter Notebooks/SupplFig7.ipynb
Supplemental Figure 8	Jupyter Notebooks/SupplFig8.ipynb
Supplemental Figure 9	Jupyter Notebooks/SupplFig9.ipynb
Supplemental Figure 10	Jupyter Notebooks/SupplFig10.ipynb
Supplemental Figure 11	Jupyter Notebooks/SupplFig11.ipynb
Supplemental Figure 12	Jupyter Notebooks/SupplFig12.ipynb

To run the notebooks and scripts you need following packages :

tensorflow, scanpy, numpy, matplotlib, scipy, wget.

Getting Started

cd code/
python DataDownloader.py
python ModelTrainer.py all

Then you can run each notebook and reproduce the results.

All datasets are available in this drive directory.