Atom, atom-type, and total nonstochastic and stochastic quadratic fingerprints: A promising approach for modeling of antibacterial activity

Yovani Marrero-Ponce, Ricardo Medina-Marrero, Francisco Torrens, Yamile Martinez, Vicente Romero-Zaldivar, Eduardo A. Castro

Producción científica: Contribución a una revistaArtículorevisión exhaustiva

85 Citas (Scopus)

Resumen

The TOpological MOlecular COMputer Design (TOMOCOMD-CARDD) approach has been introduced for the classification and design of antimicrobial agents using computer-aided molecular design. For this propose, atom, atom-type, and total quadratic indices have been generalized to codify chemical structure information. In this sense, stochastic quadratic indices have been introduced for the description of the molecular structure. These stochastic fingerprints are based on a simple model for the intramolecular movement of all valence-bond electrons. In this work, a complete data set containing 1006 antimicrobial agents is collected and presented. Two structure-based antibacterial activity classification models have been generated. The models (including nonstochastic and stochastic indices) classify correctly more than 90% of 1525 compounds in training sets. These models permit the correct classification of 92.28% and 89.31% of 505 compounds in an external test sets. The TOMOCOMD-CARDD approach, also, satisfactorily compares with respect to nine of the most useful models for antimicrobial selection reported to date. Finally, a virtual screening of 87 new compounds reported in the antiinfective field with antibacterial activities is developed showing the ability of the TOMOCOMD-CARDD models to identify new leads as antibacterial.

Idioma originalInglés
Páginas (desde-hasta)2881-2899
Número de páginas19
PublicaciónBioorganic and Medicinal Chemistry
Volumen13
N.º8
DOI
EstadoPublicada - 15 abr. 2005
Publicado de forma externa

Huella

Profundice en los temas de investigación de 'Atom, atom-type, and total nonstochastic and stochastic quadratic fingerprints: A promising approach for modeling of antibacterial activity'. En conjunto forman una huella única.

Citar esto