Fast Polypharmacy Side Effect Prediction Using Tensor Factorisation
This work addresses the critical problem of predicting adverse drug reactions for patients on multiple medications, offering a fast and accurate computational solution.
The paper tackles polypharmacy side effect prediction by demonstrating that tensor factorisation models, specifically SimplE, achieve state-of-the-art performance with median scores of 0.978 AUROC, 0.971 AUPRC, and 1.000 AP@50 across 963 side effects, while reaching 98.3% of maximum performance in just two epochs (about 4 minutes).
Motivation: Adverse reactions from drug combinations are increasingly common, making their accurate prediction a crucial challenge in modern medicine. Laboratory-based identification of these reactions is insufficient due to the combinatorial nature of the problem. While many computational approaches have been proposed, tensor factorisation models have shown mixed results, necessitating a thorough investigation of their capabilities when properly optimized. Results: We demonstrate that tensor factorisation models can achieve state-of-the-art performance on polypharmacy side effect prediction, with our best model (SimplE) achieving median scores of 0.978 AUROC, 0.971 AUPRC, and 1.000 AP@50 across 963 side effects. Notably, this model reaches 98.3\% of its maximum performance after just two epochs of training (approximately 4 minutes), making it substantially faster than existing approaches while maintaining comparable accuracy. We also find that incorporating monopharmacy data as self-looping edges in the graph performs marginally better than using it to initialize embeddings. Availability and Implementation: All code used in the experiments is available in our GitHub repository (https://doi.org/10.5281/zenodo.10684402). The implementation was carried out using Python 3.8.12 with PyTorch 1.7.1, accelerated with CUDA 11.4 on NVIDIA GeForce RTX 2080 Ti GPUs. Contact: oliver.lloyd@bristol.ac.uk Supplementary information: Supplementary data, including precision-recall curves and F1 curves for the best performing model, are available at Bioinformatics online.