Augmented three-mode MIA-QSAR modeling for a series of anti-HIV-1 compounds

Abstract

The bioactivities of a series of 2-amino-6-arylsulfonylbenzonitriles and their thio- and sulfinyl congeners have been previously modeled by using the MIA-QSAR approach and the PLS regression method [10]. The present work reports the significant improvement in the prediction ability of the Multivariate Image Analysis applied to Quantitative Structure-Activity Relationship (MIA -QSAR) model by applying Parallel Factor Analysis (PARAFAC) and Artificial Neural Network (ANN) directly to the three-way array built. This perspective represents an important advance for the accurate prediction of potential drug candidates. © 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim.