Status and Prospects of Molecular Simulations for Drug Discovery

Abstract

During the past 30 years, molecular simulations (MS) have become increasingly popular among medicinal chemists, being recognized as a promising tool to reduce drug development costs and timelines while increasing its success. From the refinement of docking experiments during the 1990’s to the contemporary possibility to produce accurate models of ligand intrinsic activity and affinity, the use of simulations in medicinal chemistry has progressed rapidly, becoming an essential tool in modern drug discovery pipelines. However, in spite of their potential, in silico methods and MS have historically been dependent on a tradeoff between computational cost and accuracy, which not uncommonly limits their impacts on drug development. In this context, this review presents a concise and up-to-date summary of the role of MS in medicinal chemistry, and examines its participation in recent advancements of both ligand-based and structure-based drug discovery, free-energy predictions, and model accuracies. In addition to offering a synopsis of the potential benefits of MS on drug discovery, this review intends to aid users to both understand the limitations of computational methods and to design more accurate experiments that, as a consequence, will have an increasing impact on the development of new therapeutics.