A Comparative Benchmarking and Evaluation Framework for Heterogeneous Network-Based Drug Repositioning Methods

Heterogeneous network-based methods are prominent in computational drug repositioning, but systematic evaluations are lacking. We conducted a comprehensive benchmarking study of 28 such methods on 11 datasets. Our evaluation framework assesses performance, scalability, and usability. The results show that methods based on matrix completion or factorization, such as HGIMC, ITRPCA, and BNNR, exhibit the best overall performance. To aid researchers, we developed HN-DREP, an online tool for viewing evaluation results, and HN-DRES, a Snakemake workflow to standardize future benchmarking.

Recommended citation: Li, Y., Yang, Y., Tong, Z. et al. (2024). "A Comparative Benchmarking and Evaluation Framework for Heterogeneous Network-Based Drug Repositioning Methods." Briefings in Bioinformatics. 25(3):bbae172.
Download Paper