A Computational Approach: Predicting iNOS Inhibition of Compounds for Alzheimer's Disease Treatment Through QSAR Modeling

Abstract

This article presents the development of a quantitative structure-activity relationship (QSAR) model for predicting the inhibitory activity of inducible nitric oxide synthase (iNOS) by specific compounds used in Alzheimer's disease treatment. iNOS is a vital enzyme involved in nitric oxide (NO) production, contributing to neuroinflammation and neuronal damage in Alzheimer's disease. The QSAR model was developed using a dataset of 90 compounds with known iNOS inhibition activity. Molecular descriptors representing the compounds’ structural and physicochemical properties were calculated and employed as input variables. Five descriptors (MATS6p, Chi1_EA(dm), Mor17 s, NsssCH, and SHED_AL) were selected as optimal for developing the classification model. The Random Forest algorithm was chosen as the classifier, implemented using WEKA software. The model underwent rigorous internal and external validation to assess its performance. The resulting QSAR model exhibited outstanding predictive capabilities with an overall accuracy of 88.8 %, a high correlation coefficient, and minimal prediction errors. It effectively forecasts iNOS inhibition activity of the chosen compounds, offering valuable insights for potential Alzheimer's disease treatments. This model significantly contributes to drug discovery, providing a rapid and cost-effective means of screening and prioritizing compounds with iNOS inhibitory potential.