Python 3.8.10
RDKit 2022.09.5
torch: 1.11.0+cu113
dgl: 0.9.1.post1
dgllife: 0.3.2
numpy: 1.23.5
scikit-learn: 1.1.3
pandas: 1.5.3
All the datasets we use are publicly available.The 'datasets' folder of this project already contains all the required datasets.
BindingDB can be obtained at https://www.bindingdb.org/bind/index.jsp;
BioSNAP and initial version DAVIS datasets can be obtained at https://github.com/kexinhuang12345/MolTrans;
Human source is at https://github.com/lifanchen-simm/transformerCPI.
python main.py --cfg "configs/DefuseDTI.yaml" --data "bindingdb" --split "random"
python main.py --cfg "configs/DefuseDTI.yaml" --data "biosnap" --split "random"
python main.py --cfg "configs/DefuseDTI.yaml" --data "human" --split "cold"
python main.py --cfg "configs/DefuseDTI.yaml" --data "human" --split "random"
python main.py --cfg "configs/DefuseDTI.yaml" --data "DAVIS" --split "random"
Dear editors and reviewers, please feel free to contact us if there are any issues during the software execution！