Suitability of GRIND-based principal properties for the description of molecular similarity and ligand-based virtual screening

Mostra el registre complet Registre parcial de l'ítem

• dc.contributor.author Durán Alcaide, Ángelca
• dc.contributor.author Zamora Rico, Ismaelca
• dc.contributor.author Pastor Maeso, Manuelca
• dc.date.accessioned 2012-07-05T06:47:44Z
• dc.date.available 2012-07-05T06:47:44Z
• dc.date.issued 2009ca
• dc.description.abstract The information provided by the alignment-independent GRid Independent Descriptors (GRIND) can be condensed by the application of principal component analysis, obtaining a small number of principal properties (GRIND-PP), which is more suitable for describing molecular similarity. The objective of the present study is to optimize diverse parameters involved in the obtention of the GRIND-PP and validate their suitability for applications, requiring a biologically relevant description of the molecular similarity. With this aim, GRIND-PP computed with a collection of diverse settings were used to carry out ligand-based virtual screening (LBVS) on standard conditions. The quality of the results obtained was remarkable and comparable with other LBVS methods, and their detailed statistical analysis allowed to identify the method settings more determinant for the quality of the results and their optimum. Remarkably, some of these optimum settings differ significantly from those used in previously published applications, revealing their unexplored potential. Their applicability in large compound database was also explored by comparing the equivalence of the results obtained using either computed or projected principal properties. In general, the results of the study confirm the suitability of the GRIND-PP for practical applications and provide useful hints about how they should be computed for obtaining optimum results.
• dc.format.mimetype application/pdfca
• dc.identifier.citation Durán A, Zamora I, Pastor M. Suitability of GRIND-based principal properties for the description of molecular similarity and ligand-based virtual screening. J Chem Inf Model. 2009;49(9):2129-38. DOI: 10.1021/ci900228xca
• dc.identifier.doi http://dx.doi.org/10.1021/ci900228x
• dc.identifier.issn 1549-9596ca
• dc.identifier.uri http://hdl.handle.net/10230/16648
• dc.language.iso engca
• dc.publisher American Chemical Society (ACS)ca
• dc.relation.ispartof Journal of Chemical Information and Modeling. 2009;49(9):2129-38
• dc.rights © 2009, American Chemical Society (ACS). This document is the unedited Author’s version of a Submitted Work that was subsequently accepted for publication in Journal of Chemical Information and Modeling, copyright © American Chemical Society after peer review. To access the final edited and published work see http://pubs.acs.org/doi/abs/10.1021/ci900228xca
• dc.rights.accessRights info:eu-repo/semantics/openAccess
• dc.subject.other Molècules -- Models -- Simulació per ordinador
• dc.title Suitability of GRIND-based principal properties for the description of molecular similarity and ligand-based virtual screeningca
• dc.type info:eu-repo/semantics/articleca
• dc.type.version info:eu-repo/semantics/submittedVersion