Rxivist logo

IHRW: An Improved Hypergraph Random Walk Model for Predicting Three-Drug Therapy

By Qi Wang, Guiying Yan

Posted 27 Feb 2021
bioRxiv DOI: 10.1101/2021.02.25.432979

Drug combination therapy is a well-established concept in the treatment of complex diseases due to its fewer side effects, lower toxicity, and better efficacy. However, it is challenging to identify efficacious drug combinations from many drug candidates. Computational models could greatly reduce the cost, but most models did not use data for more than two-drug combinations and could not predict three-drug therapy. However, three-drug combinations account for about 21% of the known combinations, which is a very important type of treatment. Here, we utilized higher-order information and developed an improved hypergraph random walk model (IHRW) for three-drug therapy prediction. This is the first method to explore the combination of three drugs. As a result, the case studies of breast cancer, lung cancer, and colon cancer showed that IHRW had a powerful ability to predict potential efficacious three-drug combinations, which provides new prospects for complex disease treatment. The code of IHRW is freely available at https://github.com/wangqi27/IHRW.

Download data

  • Downloaded 82 times
  • Download rankings, all-time:
    • Site-wide: 139,831
    • In bioinformatics: 10,787
  • Year to date:
    • Site-wide: 57,599
  • Since beginning of last month:
    • Site-wide: 63,430

Altmetric data

Downloads over time

Distribution of downloads per paper, site-wide