Doctor of Philosophy (PhD)
Natural proteins are results of evolution and they need to maintain certain thermodynamic stabilities in order to carry out their biological functions. By simulating protein evolution based on thermodynamic rules, we could reconstruct the evolution trajectory and analyze the evolutionary dynamics of a protein population, and further understand the protein sequence-structure-function relationship. In this study, we have used both a simplified lattice model and a high-resolution atomic model to simulate protein evolution processes. With the lattice model, we have investigated general theoretical questions about how protein structural designability would affect protein evolution, particularly how it would affect protein recombination and protein-ligand interactions in the evolution process. With the atomic model, we could simulate evolution processes for particular protein with different selection pressure. First, we simulated directed evolution processes and utilized such model to investigate the thermostabilization of T4 lysozyme. Second, we simulated neutral evolution processes for HIV protease, investigated its evolutionary dynamics and the possible drug-resistance mechanism in such neutral evolution. Overall, thermodynamic models can help us understand either general protein evolution dynamics or specific protein sequence-structure-function relationship in evolution.
Document Availability at the Time of Submission
Release the entire work immediately for access worldwide.
Xu, Yanlong, "Simulating protein evolution via thermodynamic models" (2006). LSU Doctoral Dissertations. 3658.