Lyman Edward, Zuckerman Daniel M
Center for Biophysical Modeling and Simulation, University of Utah, 315 S 1400 E, Rm. 2020, Salt Lake City, Utah 84112-0850, USA.
J Chem Phys. 2009 Feb 28;130(8):081102. doi: 10.1063/1.3081626.
We recently applied a multistage reweighting scheme to demonstrate the sampling of equilibrium configurational distributions of peptides from nonequilibrium, simulated annealing trajectories [E. Lyman and D. M. Zuckerman, J. Chem. Phys. 127, 065101 (2007)]. Here we demonstrate that a statistical variance reduction technique, resampling, improves the efficiency of the protocol by about a factor of 3 in a penta-alanine system. While we are not optimistic for the ultimate efficiency of purely temperature-based sampling methods, resampling-type improvements ought to find application in other (formally equivalent) nonequilibrium sampling protocols, such as Jarzynski-relation calculations and annealing-based NMR structure calculations.