A combined LS-SVM & MLR QSAR workflow for predicting the inhibition of CXCR3 receptor by quinazolinone analogs

Abstract

AnovelQSARworkflowis constructed that combines MLR with LS-SVM classification techniques for the identification of quinazolinone analogs as "active" or "nonactive" CXCR3 antagonists. The accuracy of the LS-SVM classification technique for the training set and testwas 100% and 90%, respectively. For the "active" analogs a validated MLR QSAR model estimates accurately their I-IP10 IC50 inhibition values. The accuracy of the QSAR model (R2 =0.80) is illustrated using various evaluation techniques, such as leave-one-out procedure (R2 LOO = 0.67) and validation through an external test set (R 2 pred = 0.78). The key conclusion of this study is that the selected molecular descriptors, Highest Occupied Molecular Orbital energy (HOMO), Principal Moment of Inertia along X and Y axes PMIX and PMIZ, Polar Surface Area (PSA), Presence of triple bond (PTrplBnd), and Kier shape descriptor (1κ), demonstrate discriminatory and pharmacophore abilities.