The Impact of Supervised Learning Methods in Ultralarge High-Throughput Docking

CAVASOTTO, CLAUDIO N; DI FILIPPO, JUAN I.

JOURNAL OF CHEMICAL INFORMATION AND MODELING

AMER CHEMICAL SOC

Año: 2023 vol. 63 p. 2267 - 2280

1549-9596

Structure-based virtual screening methods are nowadays one of the key pillars ofcomputational drug discovery. In recent years, a series of studies have reported docking based virtual screening campaigns of large databases ranging in the hundreds to thousandsmillions compounds, further identifying novel hits after experimental validation.As these large scale efforts are not generally accessible, machine learning-based protocolshave emerged to accelerate the identification of virtual hits within an ultra-largechemical space, reaching impressive reductions in computational time. Herein, we illustratethe motivation and the problematic behind the screening of large databases,providing an overview of key concepts and essential applications of machine learning accelerated protocols, specifically concerning supervised learning methods. We also discuss where the field stands with these novel developments, highlighting possibleinsights for future studies.