Installation¶

Installation Tested on Ubuntu 16.04, CentOS 7, MacOS catalina with Python 3.6.8 on one NVIDIA RTX 2080Ti GPU.

From Source:¶

Start by grabbing this source codes:

git clone https://github.com/juexinwang/scGNN.git
cd scGNN

Option 1 : (Recommended) Use python virutal environment with conda（https://anaconda.org/）¶

conda create -n scgnnEnv python=3.6.8 pip
conda activate scgnnEnv
pip install -r requirements.txt

If want to use LTMG (Recommended but Optional, will takes extra time in data preprocessing):

conda install r-devtools
conda install -c cyz931123 r-scgnnltmg

Option 2 : Direct install individually¶

Need to install R packages first, tested on R >=3.6.1:

In R command line:

install.packages("devtools")
library(devtools)
install_github("BMEngineeR/scGNNLTMG")

Then install all python packages in bash.

pip install -r requirements.txt

Option 3: Use Docker¶

Download and install docker.

Pull docker image gongjt057/scgnn from the dockerhub. Please beware that this image is huge for it includes all the environments. To get this docker image as base image type the command as shown in the below:

docker pull gongjt057/scgnn:code

Type docker images to see the list of images you have downloaded on your machine. If gongjt057/scgnn in the list, download it successfully.

Run a container from the image.

docker run -it gongjt057/scgnn:code /bin/bash
cd /scGNN/scGNN

Then you can proceed to the next step.