Guided structure-based ligand identification and design via artificial intelligence modelling

JUAN I. DI FILIPPO; CAVASOTTO, CLAUDIO N

EXPERT OPINION ON DRUG DISCOVERY

INFORMA HEALTHCARE

Lugar: London; Año: 2022 vol. 17 p. 71 - 78

1746-0441

Introduction:The implementations of Artificial Intelligence (AI) methodologiesapplied to drug discovery (DD) are in the rise. Several applicationshave been developed for structure-based DD, where AI methods providean alternative framework for the identification of ligands forvalidated therapeutic targets, as well as the denovodesign of ligands through generative models. Areascovered:Herein, we review contributions of the 2019-present period regardingthe application of AI methods in structure-based virtual screening(SBVS) which encompass mainly molecular docking applications -binding pose prediction and binary classification for ligand or hitidentification-, as well as denovodrug design driven by machine learning (ML) generative models, andthe validation of AI models in structure-based screening. Studies arereviewed in terms of their main objective, used databases,implemented methodology, input and output, then summarizing their keyresults.Expertopinion:It is evident that the emerging methods driven by AI will becomestandard in the DD pipeline. With the rise of new applications, moreprofound analyses regarding the validity and applicability of thesemethods have begun to appear. In the near future we expect to seemore structure-based generative models- which are scarce incomparison to ligand-based generative models-, the implementation ofstandard guidelines for validating the generated structures, and moreanalyses regarding the validation of AI methods in structure-basedDD. p { margin-bottom: 0.1in; direction: ltr; line-height: 120%; text-align: left; orphans: 2; widows: 2 }a:link { color: #0563c1 }