This is the code of paper “GATAEPR: A Graph Attention Autoencoder for Predicting Disease Progression Relationships in Cancer Patients”
Guide Steps:
1、Download source codes from github
2、Install anconda , prepare python environment and cuda environment.
A nvidia graphics card is not required, but it helps if you have one.
3、Install requires from setup.py( by conda install or pip install)
some commands:
pip install rpy2
conda install pytorch pytorch-cuda=11.6 -c pytorch -c nvidia
conda install matplotlib==3.5.1 (I found that the latest matplotlib has bugs)
4、Run codes from main.py ggae_predict function can show the result of k folds.
5、If you want to run dynoverse you need install docker(it may need 8G RAM) and r environment on your PC.
visit https://dynverse.org/ for dynoverse guidence.
If you have any questions please ask in Issues.
We are glad to answer your questions and take your suggestions.