Review of In Silico Methods for Multi-drug Combination Discovery

Combination therapy has emerged as a highly effective strategy for medicating complex diseases. With the proliferation of diverse biological data, computational methodologies have been extensively employed to pinpoint candidate drug combinations. Over the recent years, numerous models for predicting drug combinations have come to the fore, and several systematic reviews have been published on this topic. However, these reviews primarily focus on two-drug models, with limited attention given to multi-drug or high-order drug combinations. Therefore, the objective of this review is to provide a comprehensive overview of existing multi-drug prediction models. The review begins by itemizing potential data sources that may aid in the prediction of multi-drug combinations. It then summarizes the various computational models utilized for exploring multi-drug combinations. Finally, the review concludes by highlighting the key challenges and future directions for predictive multi-drug models.