README.md

Demo code for ranking protein-protein interfaces using Graph Neural Networks

This repository uses Graph Neural Networks (GNN's) to rank the docked protein complexes in the order of near-nativity to its co-crystalized native complex.

Run on Google collab:  

Dependency

• Python 3.7
  
• Jupyter notebook
  
• PyTorch 1.9.0 library
  
• NumPy
  
• Pandas
• Matplolib

Install dependencies

pip3 install -r requirements.txt

Usage

jupyter notebook

and open "Ranking protein-protein interfaces using Graph Convolution.ipynb"