CoSMIG

Communicative Subgraph Representation Learning for Multi-Relational Inductive Drug-Gene Interaction Prediction

This is the standalone code for our paper: Communicative Subgraph Representation Learning for Multi-Relational Inductive Drug-Gene Interaction Prediction

Requirements

Stable version: Python 3.7.9 + PyTorch 1.7.1+cu110 + PyTorch_Geometric 1.6.3.

Install PyTorch

Install PyTorch_Geometric

Other required python libraries: numpy, scipy, pandas, h5py, networkx, tqdm etc.

Also you can install the required packages follow there instructions (tested on a linux terminal):

conda env create -f environment.yaml

Datasets

Please Contact us (raojh6@mail2.sysu.edu.cn) to obtain the Data (from DrugBank and DGIdb) and Splits.

Statistic of DGI Dataset

Usages

For training on DrugBank on the transductive scenario:

CUDA_VISIBLE_DEVICES=0 python main.py --data-name DrugBank --testing --dynamic-train --dynamic-test --dynamic-val --save-results --max-nodes-per-hop 200

For training on DGIdb on the inductive scenario:

CUDA_VISIBLE_DEVICES=0 python main.py --data-name DGIdb --testing --mode inductive --dynamic-train --dynamic-test --dynamic-val --save-results --max-nodes-per-hop 200

More parameters could be found by:

python main.py -h

Reference

If you find the code useful, please cite our paper.

@inproceedings{cosmig,
  title     = {Communicative Subgraph Representation Learning for Multi-Relational Inductive Drug-Gene Interaction Prediction},
  author    = {Rao, Jiahua and Zheng, Shuangjia and Mai, Sijie and Yang, Yuedong},
  booktitle = {Proceedings of the Thirty-First International Joint Conference on
               Artificial Intelligence, {IJCAI-22}},
  publisher = {International Joint Conferences on Artificial Intelligence Organization},
  editor    = {Lud De Raedt},
  pages     = {3919--3925},
  year      = {2022},
  month     = {7},
  note      = {Main Track},
  doi       = {10.24963/ijcai.2022/544},
  url       = {https://doi.org/10.24963/ijcai.2022/544},
}

Contact

Jiahua Rao (raojh6@mail2.sysu.edu.cn) and Yuedong Yang (yangyd25@mail.sysu.edu.cn)