9509100 Madrid This research will explore the convergence of comparative modeling, molecular dynamics and simulated annealing techniques, as a function of the degree of homology of the protein structures used as templates. Comparative modeling techniques will be used to predict the structure of several proteins from weakly and highly homologous sequences. The predicted structures will be subject to molecular dynamics and simulated annealing protocols, to study the efficiency and accuracy with which these techniques bring the predicted structures into better agreement with those experimentally determined. Novel protocols will be tested and compared, including molecular dynamics in the picosecond and nanosecond time scales, sequential simulated annealing and parallel Monte Carlo simulated annealing. The accuracy of the calculations depend, among other factors, on the approximations used to treat the electrostatic interactions and to simulate the solvent effects. A novel treatment of the electrostatic interactions, the Particle Mesh Ewald protocol, will be applied. Both crystal and solution structures will be simulated, for comparison purposes. Proteins will be selected from the Brookhaven Protein Data Bank (PDB) based on the existence of a family of experimentally determined structures from different species. They will include phospholipase, lysozyme and myoglobin. The calculated protein structures will be compared to the corresponding X-ray crystallography and NMR-determined structures from the PDB. %%% This research will explore the convergence of comparative modeling, molecular dynamics and simulated annealing techniques, as a function of the degree of homology of the protein structures used as templates. Comparative modeling techniques will be used to predict the structure of several proteins from weakly and highly homologous sequences. This will be valuable in determining the confidence level that can be placed on predicted structures. The knowledge gained from this research can be directly applied to the prediction of unknown protein structures and mutants, the refinement of experimentally determined structures, and to the rational design of macromolecules to be prepared by site-directed mutagenesis techniques. ***
