Comparative QSAR based on neural networks for the anti-HIV activity of HEPT derivatives.

摘要: Among the non-nucleoside reverse transcriptase inhibitors, 1-[2-hydroxyethoxy-methyl]-6-(phenylthio) thymine (HEPT) derivatives have proved to be potent and selective inhibitors of human immunodeficiency virus (HIV-1). They are able completely suppress replication in cell cultures. The quantitative structure-activity relationships (QSAR) try describe association between biological activities a group congeners their molecular descriptors. In this paper, recent works on application neural networks (NN) multiple regression analyses structure-anti-HIV activity HEPT reviewed. NN origins efforts reproduce computer models information processing that takes place brain. found wide variety fields, such as image analysis facial features, stock market predictions, etc. Application methods problems chemistry biochemistry has rapidly gained popularity years. We briefly methodology for designing QSAR estimating performances, apply approach prediction anti-HIV HEPT. predictive power used is compared with other statistical methods.