Comparative molecular field analysis (CoMFA) and comparative molecular similarity indices analysis (CoMSIA) three dimensional structure-activity relationship (3D-QSAR) studies were conducted on a series of novel triazole antifungal compounds. For the CoMFA study, two different pharmacophoric conformations were compared and conformation generated from our previous homologous enzyme-inhibitor docking model got better results. Variation of grid spacing was used during the optimization of the CoMFA model. For the CoMSIA study, the influence of the combination of different field types was evaluated and the best combination was considered to be of steric, electrostatic, hydrophobic and H-bonding acceptor fields. Variation of grid spacing and attenuation factor was used to get the best CoMSIA model. The resulting CoMFA and CoMSIA models had a cross validated coeffiecient q² of 0.718 and 0.655 respectively, which showed strong predictive ability on both test set and training set. The 3D contour maps of CoMFA and CoMSIA provided smooth and interpretable explanation of the structure-antifungal activity relationship for the compounds. The analysis of the 3D contour plots permitted interesting conclusions about the effects of different substituent groups at different position of the phenyl group on the antifungal activity, which will guide the design of novel antifungal compounds with relatively high activity.

Key words: triazole derivatives, antifungal, CoMFA, CoMSIA, 3D-QSAR