A review of deep learning approaches for drug synergy prediction in cancer

Abstract

Synergistic drug combinations enhance cancer treatment by improving efficacy and reducing toxicity. With advances in artificial intelligence and large-scale datasets, deep learning has become central to anti-cancer drug synergy prediction. This review summarizes classical and emerging deep learning models from single-task learning and multi-task learning perspectives, discusses data and technical challenges, and highlights future directions for advancing computational drug synergy prediction.