/DeepTGIN

DeepTGIN: a novel hybrid multimodal approach using transformers and graph isomorphism networks for protein-ligand binding affinity prediction

Primary LanguagePython

DeepTGIN

DeepTGIN: a novel hybrid multimodal approach using Transformers and graph isomorphism networks for protein-ligand binding affinity prediction

conda create -n DeepTGIN poython=3.8

Data

We use PDBBind2020 as our dataset，http://pdbbind.org.cn/ The code for data preprocessing is create_data.py

Train

If you want to train the model, run python main.py

Cite

If our work is helpful to you, we encourage you to cite our paper: