3D-QSAR and pharmacophore model study on aryl diphenolic azoles as estrogen receptor-β ligands

Abstract

Pharmacophore and three-dimensional quantitative structure–activity relationship (3D-QSAR) model of 106 aryl diphenolic azoles as estrogen receptor-β ligands were proposed. The best pharmacophore model, five-site model including two hydrogen bond donors and three aromatic rings, has been established. The biological activity values (pIC50) and classification (active or inactive) of compounds were predicted by 3D-QSAR model. The model correlation coefficient (R 2) is 0.9623. The classification accuracies in predicting for training, test, and total datasets are 97.5, 84.6, and 94.3 %, respectively. The 3D-QSAR model maps revealed that hydrogen bond donor, hydrophobic, and electron-withdrawing fields would be the essential factors for designing new ligands with higher biological activity. The results indicate that the combination of pharmacophore and 3D-QSAR analyses could be a powerful modeling tool for finding novel estrogen receptor-β ligands.