OPTIMIZATION METHODS FOR CONFORMATIONAL SAMPLING USING A BOLTZMANN-WEIGHTED MEAN-FIELD APPROACH

An optimization protocol is proposed that combines a mean field simula tion approach with Boltzmann-weighted sampling. This is done by includ ing Boltzmann probabilities of multiple conformations in the optimizat ion procedure. The method is demonstrated on a simple model system and on the side-chain conformations of phenylalanines in a small hexapept ide. For comparison, calculations were performed using classical stoch astic dynamics simulations [M. Saunders, K. N. Houk, Y. Wu, C. Still, M. Lipton, G. Chang, and W. C. Guida (1990), Journal of the American C hemical Society, Vol. 12, pp. 1419], iterative optimization of probabi lities on a fixed set of basis conformations [P. Koehl and M. Delaure (1994), Journal of Molecular Biology, Vol. 239, pp. 249-275], and simu lations with locally enhanced sampling [A. Roitberg and R. Elber (1991 ), Journal of Chemical Physics, Vol. 95, pp. 9277-9287]. Although appr oximations are used in our method, the results may be more physically meaningful than those of the other procedures discussed. Furthermore, the approximate Boltzmann distribution allows generalization of the re sults. (C) 1996 John Wiley & Sons, Inc.