ISSN print edition: 0366-6352 ISSN electronic edition: 1336-9075 Registr. No.: MK SR 9/7 Published monthly

A new, fully validated and interpreted quantitative structure-activity relationship model of p-aminosalicylic acid derivatives as neuraminidase inhibitors

Ana Paula Hartmman, Daniela Hartmman Jornada, and Eduardo Borges de Melo

Theoretical Medicinal and Environmental Chemistry Laboratory (LQMAT), Department of Pharmacy, Western Paraná State University (Unioeste), 2069 Universitária St, 85819110, Cascavel, State of Paraná, Brazil

E-mail: eduardo.melo@unioeste.br

Abstract: A multivariate QSAR study with a set of 34 p-aminosalicylic acid derivatives, described as neuraminidase inhibitors of the H1N1 viruses, is presented in this work. The variable selection was performed with the Ordered Predictors Selection (OPS) algorithm and the model was built with the Partial Least Squares (PLS) regression method. Leave-N-out cross-validation and y-randomization tests showed that the model was robust and free from chance correlation. The external predictive ability was superior to the 3D-QSAR model previously published. Moreover, it was possible to perform a mechanistic interpretation, where the descriptors referred directly to the mechanism of interaction with the neuraminidase.

Keywords: QSAR – H1N1 – OPS – PLS – model validation – mechanistic interpretation

Full paper is available at www.springerlink.com.

DOI: 10.2478/s11696-013-0321-0

Chemical Papers 67 (5) 556–567 (2013)