Convenient QSAR model for predicting the complexation of structurally diverse compounds with beta-cyclodextrins

• Tipo: Artigo em Revista Científica Internacional
• Data: 2009

Autores

Alfonso Perez Garrido (Autor) Outra; Aliuska M Morales Helguera (Autor) Outra; Adela A Abellan Guillen (Autor) Outra; Natalia N D S Cordeiro (Autor) FCUP; Amalio G Garrido Escudero (Autor) Outra

Revista

Título: Bioorganic and Medicinal Chemistry

Vol. 17

Páginas: 896-904

ISSN: 0968-0896

Editora: Elsevier

Indexação

ISI Web of Knowledge - 15 Citações

Scopus - 24 Citações

Classificação Científica

FOS: Ciências exactas e naturais > Química

Outras Informações

ID Authenticus: P-003-NRR

DOI: 10.1016/j.bmc.2008.11.040

Abstract (EN): This paper reports a QSAR study for predicting the complexation of a large and heterogeneous variety of substances (233 organic compounds) with beta-cyclodextrins (beta-CDs). Several different theoretical molecular descriptors, calculated solely from the molecular structure of the compounds under investigation, and an efficient variable selection procedure, like the Genetic Algorithm, led to models with satisfactory global accuracy and predictivity. But the best-final QSAR model is based on Topological descriptors meanwhile offering a reasonable interpretation. This QSAR model was able to explain ca. 84% of the variance in the experimental activity, and displayed very good internal cross-validation statistics and predictivity on external data. It shows that the driving forces for CD complexation are mainly hydrophobic and steric (van der Waals) interactions. Thus, the results of our study provide a valuable tool for future screening and priority testing of beta-CDs guest molecules.

Idioma: Inglês

Tipo (Avaliação Docente): Científica

Nº de páginas: 9