Grandline

Graph convolutional neural network classification model for integrating biological network and gene expression (Grandline). It is a framework based on graph convolutional neural network for predicting phenotype based on an integration of gene expression data and a protein-protein interaction network. It can also identify important subnetworks that are critical for phenotype prediction in each sample.

How to use Grandline

git lfs clone https://github.com/BioML-CM/Grandline
conda create -n grandline python=3.7
conda activate grandline
cd Grandline
pip install -r requirements.txt
nohup jupyter notebook --port 8888 &

Next, open tutorial.ipynb on your browser.

To install git lfs on Ubuntu

Alternatively, you may download the example dataset directly from github and add to data directory.

curl -s https://packagecloud.io/install/repositories/github/git-lfs/script.deb.sh | sudo bash
sudo apt-get install git-lfs
git lfs install

For other operating systems: https://github.com/git-lfs/git-lfs/wiki/Installation