Efficiency of simulated annealing for peptides with increasing geometricalrestrictions

Simulated annealing (SA) is a popular global minimizer that can convenientl y be applied to complex macromolecular systems. Thus, a molecular dynamics or a Monte Carlo simulation starts at high temperature, which is decreased gradually, and the system is expected to reach the low-energy region on the potential energy surface of the molecule. However, in many cases this proc ess is not efficient. Alternatively, the low-energy region can be reached m ore effectively by minimizing the energy of selected molecular structures g enerated along the simulation pathway. The efficiency of SA to locate energ y-minimized structures within 5 kcal/mol above the global energy minimum is studied as applied to three peptide models with increasing geometrical res trictions: (1) The Linear pentapeptide Leu-enkephalin described by the ECEP P potential, (2) a cyclic hexapeptide described by the GROMOS force field e nergy E-GRO alone, and (3) the same cyclic peptide with E-GRO combined with a restraining potential based on 31 proton-proton restraints obtained from nuclear magnetic resonance (NMR) experiments. The efficiency of SA is comp ared to that of the Monte Carlo minimization (MCM) method of Li and Scherag a, and to our local torsional deformations (LTD) method for the conformatio nal search of cyclic molecules. The results for the linear peptide show tha t SA provides a relatively weak guidance towards the most stable energy reg ion; as expected, this guidance increases for the cyclic peptide and the cy clic peptide with NMR restraints. However, in general, MCM and LTD are sign ificantly more efficient than SA as generators of low-energy minimized stru ctures. This suggests that LTD might provide a better search tool than SA i n structure determination of protein regions for which a relatively small n umber of restraints are provided by NMR. (C) 1999 John Wiley & Sons, Inc.