This is a Pytorch implementation of
SS-GNN, a simple-structured GNN model for drug-target binding affinity (DTBA) prediction as described in the following paper:
The SS-GNN defines the prediction of DTBA as a regression task, in which the model’s input is the drug-target representation, and the output is a continuous value representing the binding affinity score between the drug and the target protein. The overall architecture of the SS-GNN is shown in the figure below.
-
Setup
Necessary packages should be installed to run the SS-GNN model. Dependecies:
- python >= 3.7
- Pytorch (>=1.6.0),
- numpy,
- scipy,
- scikit-learn.
-
Datasets
We adopt the PDBbind dataset v2019 for experiments and employ two test sets (the v2016 and v2013 core sets) to test the performance of SS-GNN.
-
Train the model
Use the
train.pyscript to train the model.
When using this project in your research, please cite:
Zhang, S., Jin, Y., Liu, T., Wang, Q., Zhang, Z., Zhao, S., & Shan, B. (2023).
SS-GNN: A Simple-Structured Graph Neural Network for Affinity Prediction.
ACS Omega, 8(25), 22496–22507.
https://doi.org/10.1021/acsomega.3c00085