EFFICIENCY OF MONTE-CARLO MINIMIZATION PROCEDURES AND THEIR USE IN ANALYSIS OF NMR DATA OBTAINED FROM FLEXIBLE PEPTIDES

The Monte Carlo minimization (MCM) method of Li and Scheraga is an eff icient tool for generating low energy minimized structures of peptides , in particular the global energy minimum (GEM). In a recent article w e proposed an enhancement to MCM, called the free energy Monte Carlo m inimization (FMCM) procedure. With FMCM the conformational search is c arried out with respect to the harmonic free energy, which approximate s the free energy of the potential energy wells around the energy mini mized structures (these wells are called localized microstates). In th is work we apply both methods to the pentapeptide Leu-enkephalin descr ibed by the potential energy function ECEPP, and study their efficienc y in identifying the GEM structure as well as the global harmonic free energy (GFM) structure. We also investigate the efficiency of these m ethods to generate localized microstates, which pertain to different e nergy and harmonic free energy intervals above the GEM and GFM, respec tively. Such microstates constitute an important ingredient of our sta tistical mechanical methodology far analyzing nuclear magnetic resonan ce data of flexible peptides. Aspects of this methodology related to t he stability properties of the localized microstates are examined. (C) 1997 by John Wiley & Sons, Inc.