HNUBioinformatics / Hts

Hts

A Granularity-level Information Fusion Strategy on Hypergraph Transformer for Predicting Synergistic Effects of Anticancer Drugs

HypertranSynergy is a hypergraph learning network model used to predict the synergistic effect of anticancer drugs. It mainly includes two parts: CIE is a coarse-grained information extraction module based on hypergraph transformer, and FIE is a fine-grained information extraction module based on attention network. By coupling the two parts into HypertranSynergy, you can make the embeddings contain rich grain-level information to make prediction.

Requirements

• Python 3.7 or higher
• PyTorch 1.8.0 or higher
• deepchem 2.5.0
• rdkit 2022.9.5
• torch-geometric 2.1.0

Data

• cell line
  • https://pan.baidu.com/s/1inc_gzfEeQRbieyAZhD7jQ 提取码：K527
  • https://pan.baidu.com/s/1uliG6nxFbo_hAAiJuuz9rA 提取码：K529
• drug
  • https://pan.baidu.com/s/1d5j1VYp2WqlnpkRgLALqPQ 提取码：K528
  • https://pan.baidu.com/s/1LnSmgc5ceFFWs1yr8jYB6g 提取码：K529
• synergy
  • https://pan.baidu.com/s/1wG4NQqnm0c88tRTX3vuplg 提取码：K529
  • https://pan.baidu.com/s/1UMCkKUNqEdoTl1UHw-sqcg 提取码：K529

Running the Code

• Running the "main_.py" for the classification task.
• Running the "main_reg.py" for the regression task.

Others

updating...