發展三維定量構效關係組合模型架構用於搜尋與優化標靶蛋白質抑制劑基於Pharmacophore, CoMFA 和 CoMSIA 電腦技術

Abstract

Quantitative Structure Activity Relationships (QSAR) is an important technique in the rational drug design, which was used to build computational models to find a statistically significant correlation between the receptor and inhibitors. There are two mainstream of 3D-QSAR technologies, namely Comparative Molecular Field Analysis (CoMFA)/ Comparative Molecular Similarity Index Analysis (CoMSIA) and Pharmacophore. Most significant function of pharamcophore model is to use 3D screen to recognize the related target protein inhibitors. However, the number of pharmacophore features was restricted as five chemical features at maximum, and which could not describe the 3D space limitation of the binding site. This restriction induces incompletely describing the chemical features of inhibitors. Contrastingly, the other two models, CoMFA and CoMSIA were not suitable to search 3D databases, but can easily be used to modify the molecule structure optimization and describe the limit range of molecule weights. Additional, CoMFA and CoMSIA models use contours to describe the chemical features of inhibitor. The number of contours was not restricted, that could reflect the chemical features of inhibitor. Therefore, CoMFA and CoMSIA models could provide better predication ability to predict the bioactivity. According to above characters, we prefer to combine these two different technologies. We propose a 3D-QSAR combination modeling approach to solve two 3D-QSAR technical shortcomings of each other. Our combination approach could provide a valuable tool in the design of new leads with desired biological activity by virtual screening.