This repository is the official implementation of Spatial Graph Attention Network (sGAT) within the paper Spatial Graph Attention and Curiosity-driven Policy for Antiviral Drug Discovery. Our policy networks built upon sGAT can be found here.
conda config --append channels conda-forge
conda create -n sgat-env --file requirements.txt
conda activate sgat-env* make sure to install the right versions for your toolkit
Once the environment is set up, the function call to train & evaluate sGAT is:
./main.sh &A list of flags may be found in main.sh and main.py for experimentation with different network parameters. The run log and models are saved under *artifact_path*/saves, and the tensorboard log is saved under *artifact_path*/runs.
A trained sGAT model on a sub-dataset of molecules and scores for docking in the catalytic site of NSP15 can be found here.
Contributions are welcome! All content here is licensed under the MIT license.