Premium
Off‐lattice Monte Carlo method with constraints: Long‐time dynamics of a protein model without nonbonded interactions
Author(s) -
Knapp E.W.,
IrgensDefregger A.
Publication year - 1993
Publication title -
journal of computational chemistry
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.907
H-Index - 188
eISSN - 1096-987X
pISSN - 0192-8651
DOI - 10.1002/jcc.540140106
Subject(s) - monte carlo method , autocorrelation , statistical physics , degrees of freedom (physics and chemistry) , molecular dynamics , lattice (music) , algorithm , chemistry , physics , computer science , thermodynamics , computational chemistry , mathematics , statistics , acoustics
A method is proposed to perform computer simulations of protein dynamics in the long‐time regime. The method is based upon a Monte Carlo technique. The only molecular degrees of freedom considered are bond rotations. All other degrees of freedom including the amide plane torsions are kept rigid. These constraints approximately account for all interactions related to chemical bonding. An individual Monte Carlo step adopts the Go and Scheraga algorithm where local conformational changes in a small window of the protein backbone are performed. By using correlated rotations, the conformation of residues outside the window remains invariant. To test the reliability of the method, the nonbonded interactions are turned off in the present application. Exact statistical averages are compared with values obtained from data of computer simulation involving 2 × 10 6 scans of the window along the protein backbone. Time is related to the number of scans of the window along the protein backbone. End‐to‐end distance autocorrelation functions decay to 1/ e of its initial value in about 10 3 –10 4 scans of the window algorithm. Time decay follows a stretched exponential Kohlrausch decay law. © 1993 John Wiley & Sons, Inc.