| Molecular docking is an important tool in structure-based drug design. As more and more crystal structures are resolved, molecular docking methods have been extensively used in drug design. Meanwhile, some shortages or limitations in the present molecular docking methods have emerged, such as insufficiency of molecular conformational searching, low accuracy of scoring functions, negligence of solvent effect and protein flexibility in docking methods. Among them, how to properly deal with the protein flexibility is the most intractable issue of developing molecular docking methods. In view of this, taking the conformational changes of proteins before and after binding with ligands as the judgment criterion, this thesis performed an intensive clustering analysis on the present available protein crystal structures. It provides reliable support for drug design with considering protein flexibility. In addition, we selected some side chain-flexible proteins as a test set and performed an accuracy assessment of molecular docking methods with considering side chains flexibility. These include our in-house method iFitDock and the other two popular molecular docking methods (AutoDock and Induced-Fit Docking). The results indicate that iFitDock has an obvious advantage for predicting native binding poses of proteins and ligands over the other two molecular docking tools, and prove that iFitDock is a promising molecular docking tool for drug discovery and design. Moreover, it provides theoretical foundation for developingt and perfecting molecular docking methods which considers protein flexibilityAs an important post-translational modification enzyme, Farnesyltransferase (FTase) plays a critical role in membrane location and generation of biological effect and becomes an significant target in cancer treatment. We have identified a series of inhibitors against FTase by using virtual screening in our previous studies. In this thesis, focusing on the amidodiphenyl ethers of FTase inhibitors identified by us, we performed a structure-activity relationship (SAR) study on them by using molecular docking simulations. Based on the SAR model, chemical synthesis technology and bioassays, we proposed a structural modification strategy and obtained a series of novel inhibitors with high inhibitory activities against FTase. |
