This research is directed toward developing greatly improved methods for sampling phase space in molecular systems having multiple conformations which are separated by significant energy barriers. The work will be important in converging free energy simulations of complex systems. The method we are exploring mixes stochastic dynamics (SD) and Metropolis Monte Carlo (MC) methods. The former method provides efficient local sampling while the MC facilitates transitions between the various conformational minima. We have shown that the mixed method is significantly more efficient than either method individually. Recently, we have developed an enhancement of the MC part of the algorithm in which directly jumping between conformational minima is allowed. This further increases the sampling efficiency by several orders of magnitude. In addition to algorithm development, we continue to develop potential functions and to compare our calculations with experiment.
Showing the most recent 10 out of 22 publications