Accelerating pharmaceutical cocrystal discovery via in silico screening: prediction of formation and properties

Cocrystallization has emerged as a versatile and effective strategy to modify physicochemical properties of drugs in molecular solid-state design. However, the selection of suitable cocrystal coformers remains largely guided by empirical knowledge and trial-and-error. In recent years, in silico virtual screening methods, including knowledge-based approaches, physics-based approaches and machine learning, have been recognized as effective tools for addressing these challenges. This review provides a comprehensive overview of the theoretical principles, applications and recent advances in in silico virtual screening methods for predicting cocrystal formation and properties. The advantages and limitations of current methods are crucially analyzed, and promising directions for future research are outlined. These efforts are expected to facilitate more reliable applications of in silico tools in the rational design of pharmaceutical cocrystals.