Integrating concept of pharmacophore with Graph Neural Networks for chemical property prediction and interpretation

This repository contains the source code and the data.

RG-MPNN

Setup and dependencies

Dependencies:

• python 3.7
• pytorch = 1.7.1
• torch-cluster = 1.5.9
• torch-geometric = 1.7.2
• torch-scatter = 2.0.7
• torch-sparse = 0.6.9
• torch-spline-conv = 1.2.1
• RDkit = 2021.03.3
• numpy
• pandas

Data sets

The data sets are provided as .csv files in a directory called 'data', including benchmark datasets and kinase datasets used in this work.

---

Using

1.MyNet_Classification generates input, train and test classification models. For example,

python MyNet_Classification.py \
    --epochs 100 \
    --dataset BACE \
    --model RGNN 

2.MyNet_Regression generates input, train and test regression models. For example,

python MyNet_Regression.py \
    --epochs 100 \
    --dataset Lipo \
    --model RGNN

---

Author

Yue Kong

Aixia Yan

Citation