Protein Structure Of Computer Simulation And Chemical Shift Analysis Of The New Method

MD simulation is one of the important theoretical tool to study molecular systems at atomic level. With the progress of empirical force field, sampling methods and computer hardware, larger molecular systems can be simulated and more precise results would be obtained currently than before. However, there are still two issues to be solved: sampling efficiency and the accuracy of empirical force fields.Because of the low sampling efficiency of classic MD simulations, a large number of algorithms have been developed to enhance the sampling efficiency. To calculate conformational free energies of molecular systems, highly effective sampling in conformational space is required. But, well-converged free energy surface can be calculated by using previous algorithms in most cases. In Chapter 2, a new enhanced sampling algorithm has been presented, which integrates multiple replica simulation and a constrained update of umbrella potential into conventional adaptive umbrella sampling algorithm. By using this algorithm, well-converged free energy surface of alanine decapeptide has been obtained and the relative conformational free energy between α-helical and β-hairpin conformations of alanine decapeptide has also been derived. In Chapter3, a novel Hamiltonian replica exchange method is proposed, in which a series of umbrella potentials derived from the Ramachadran free energy surface of alanine dipeptide under specific force field are added to each pair of backbone dihedrals to compensate for transition barriers on Ramachandran map. And we deduced the generalized weighted histogram analysis equations to calculate free energy surfaces. This method has been applied to the two small...