SAnDReS 2.0: Development of machine-learning models to explore the scoring function space.

Autor: de Azevedo WF Jr; Department of Physics, Institute of Exact Sciences, Federal University of Alfenas, Alfenas, Brazil., Quiroga R; Instituto de Investigaciones en Fisicoquímica de Córdoba (INFIQC), CONICET-Departamento de Química Teórica y Computacional, Facultad de Ciencias Químicas, Universidad Nacional de Córdoba, Ciudad Universitaria, Córdoba, Argentina., Villarreal MA; Instituto de Investigaciones en Fisicoquímica de Córdoba (INFIQC), CONICET-Departamento de Química Teórica y Computacional, Facultad de Ciencias Químicas, Universidad Nacional de Córdoba, Ciudad Universitaria, Córdoba, Argentina., da Silveira NJF; Laboratory of Molecular Modeling and Computer Simulation, Federal University of Alfenas, Alfenas, Brazil., Bitencourt-Ferreira G; Pontifical Catholic University of Rio Grande do Sul, Porto Alegre, Brazil., da Silva AD; Programa de Pós-Graduação em Tecnologias da Informação e Gestão em Saúde, Universidade Federal de Ciências da Saúde de Porto Alegre, Porto Alegre, Brazil., Veit-Acosta M; Western Michigan University, Kalamazoo, Michigan, USA., Oliveira PR; School of Arts, Sciences and Humanities, University of São Paulo, São Paulo, Brazil., Tutone M; Dipartimento di Scienze e Tecnologie Biologiche Chimiche e Farmaceutiche (STEBICEF), Università di Palermo, Palermo, Italy., Biziukova N; Institute of Biomedical Chemistry, Moscow, Russia., Poroikov V; Institute of Biomedical Chemistry, Moscow, Russia., Tarasova O; Institute of Biomedical Chemistry, Moscow, Russia., Baud S; Laboratoire SiRMa, UMR CNRS/URCA 7369, UFR Sciences Exactes et Naturelles, Université de Reims Champagne-Ardenne, CNRS, MEDYC, Reims, France.
Jazyk: angličtina
Zdroj: Journal of computational chemistry [J Comput Chem] 2024 Oct 15; Vol. 45 (27), pp. 2333-2346. Date of Electronic Publication: 2024 Jun 20.
DOI: 10.1002/jcc.27449
Abstrakt: Classical scoring functions may exhibit low accuracy in determining ligand binding affinity for proteins. The availability of both protein-ligand structures and affinity data make it possible to develop machine-learning models focused on specific protein systems with superior predictive performance. Here, we report a new methodology named SAnDReS that combines AutoDock Vina 1.2 with 54 regression methods available in Scikit-Learn to calculate binding affinity based on protein-ligand structures. This approach allows exploration of the scoring function space. SAnDReS generates machine-learning models based on crystal, docked, and AlphaFold-generated structures. As a proof of concept, we examine the performance of SAnDReS-generated models in three case studies. For all three cases, our models outperformed classical scoring functions. Also, SAnDReS-generated models showed predictive performance close to or better than other machine-learning models such as K DEEP , CSM-lig, and Δ Vina RF 20 . SAnDReS 2.0 is available to download at https://github.com/azevedolab/sandres.
(© 2024 Wiley Periodicals LLC.)
Databáze: MEDLINE
Externí odkaz: