VeloVGI

Introduction

Graph Variational Autoencoder for scRNA-seq velocity

Installation

You are suggested to reproduce the conda environment with velovgi.yml using the following command :

conda env create -f velovgi.yml 

You can get the packages also by pip command:

pip install velovgi

The main packages version are as following :

# Optimal Transportation
pot = 0.9.0

# Deeplearning 
torch==1.12.1
pytorch-lightning==1.7.7
torch-geometric==2.0.1

# scRNA-seq 
scanpy==1.9.3
scvi-tools==0.19.0

# RNA velocity 
scvelo==0.2.5
velovi==0.2.0

Quick start

We provide a jupyter notebook for users to quickly understand the use of the tool and the output results.

Reference

The paper of the work is under review. You can cite our work with the following methods.