| With the development of biology technology, more and more protein structures are measured out. The spatial structure of a protein molecule is the foundation of researching its function. As an important part of a protein molecule, the molecular surface plays a significant role in the visualization of the molecular structure and docking. Therefore, studying protein molecules by a computation method has become an important topic in the field of molecular biology.This thesis focuses on the construction of a protein molecule surface model and the application of molecule energy distribution to surface feature analysis. The work consists of the following there aspects:1) The original Contour-Buildup algorithm is improved by combining the Extraction Surface Atoms algorithm with the Contour-Buildup algorithm. Based on the extracted surface atoms, the speed and efficiency of calculating the solvent-excluded surface is significantly improved. During the process of triangulating the SES, we improve the triangular method of convex spherical patches by using intermediate results of Contour-Buildup algorithm. Consequently, saddle patches and convex spherical patches can be connected seamlessly without spending any extra calculation time. As a result, we can obtain a uniform and seamless triangular mesh model for a protein molecule.2) A new local rendering method is proposed when generating the surface triangulation mesh of a docking compound molecule. This method can improve the computing speed and save time, because it only recalculates the molecular surface of docking parts rather than the entire surface of the new compound.3) Through analyzing the distribution of electrostatic potential energy at the molecular surface, it is revealed that the characteristics of the energy distribution in the different feature areas of the molecular surface. These characteristics can help identify the molecular surface feature areas, thereby can further help determine the position of the molecular docking. |
PDF Full Text Request