Predictive QSAR Modeling for the Successful Predictions of the ADMET Properties of Candidate Drug Molecules

Mahmud   Tareq Hassan   Khan; Ingebrigt      Sylte; Mahmud   Tareq Hassan   Khan; Ingebrigt      Sylte

Abstract

Chemical breakthrough generates large numbers of prospective drug molecules; the use of ADMET (absorption, distribution, metabolism, excretion and toxicity) properties is flattering progressively more imperative in the drug discovery, assortment, development and promotion processes. Due to the inauspicious ADMET properties a huge amount of molecules in the development stage got failure. In the past years several authors reported that it possible to do some prediction of the ADMET properties using the structural features of the molecules, suing several approaches. One of the most important approaches is QSAR modeling of the data derived from their activity profiles and their different structural features (i.e., quantitative molecular descriptors). This review is critically assessing some of the most important issues for the effective prediction of ADMET properties of drug candidates based on QSAR modeling approaches.

Keywords: ADMET prediction, QSAR modeling, molecular descriptor, multiple linear regressions, machine learning, artificial neural network