QSAR study for the prediction of IC50 and Log P for 5-N-Acetyl-Beta-DNeuraminic Acid structurally similar compounds using stepwise (multivariate) linear regression

Abstract

Multi-parametric Quantitative structure activity relationship (QSAR) study has been developed for 110 training compounds and 50 test compounds structurally similar to 5-N-ACETYL-BETA-D-NEURAMINIC ACID as inhibitors for Clostridium tetani. Stepwise (multi-parametric) Linear Regression QSAR models for biological activity of half maximal inhibitory concentration (IC50) and log P for octanol/water (Log P) were created with 16 different descriptors. The predictive capability of the QSAR models were evaluated by r2 , q2 LMO(TestSet) , q2LOO(TestSet) , q2BOOT(TestSet). The comparison of various external validation reveals identical q2 LMO(TestSet) , q2 LOO(TestSet) and q2 BOOT(TestSet) for IC50 (0.98), and Log P(0.7) which demonstrates the high robustness and real predictive power of IC50 and Log P model. LMO-Leave many out, LOO-Leave one out, BOOT-bootstrapping