Comparison Between 2D and 3D‐QSAR Approaches to Correlate Inhibitor Activity for a Series of Indole Amide Hydroxamic Acids

Abstract

AbstractThe well‐expressed inhibitor activity of 36 indole amide hydroxamic acids to Histone Deacetylases (HDACs) was analyzed by CODESSA PRO software and the Best Multiple Linear Regression (BLMR) algorithm, to provide a reliable 2D‐QSAR model from a set of more than 800 descriptors. Concurrently, Chem‐X (version 1994) software was used to develop a corresponding 3D‐QSAR model; the steric and electrostatic interactions between a fictitious probe atom “H+” and a set of aligned molecules were evaluated using the CoMFA approach as implemented in Chem‐X. A Partial Least Squares (PLS) procedure generated the principal components needed to build a 3D‐QSAR model. The Weighted Least Squares (WLS) method generated an R2 map of the regions around the molecules important for steric and electrostatic interactions. Enzyme–inhibitor docking calculations were carried out to combine and compare with the results obtained by CoMFA. New potentially active structures were proposed.