  Institute_ |
 Group_ |
| Institute_ |
Group_ |
Helmut Grubmüller and Paul Tavan
Institut für Medizinische Optik, Theoretische Biophysik
Ludwig-Maximilians-Universität München
Theresienstraße 37, D-80333 München, Germany
email: Helmut.Grubmueller@physik.uni-muenchen.de
s
have been carried out on a simplified protein model.
Despite its simplified structure, that model
exhibits properties similar to those of more realistic protein models.
In particular, the model was found to undergo
transitions between conformational substates at a time scale of several
hundred picoseconds.
The computed trajectories turned out to be sufficiently long
as to permit a statistical analysis of that conformational
dynamics.
To check whether effective descriptions neglecting memory effects can reproduce the observed conformational dynamics, two stochastic models were studied: a one-dimensional Langevin effective potential model derived by elimination of sub-picosecond dynamical processes could not describe the observed conformational transition rates. In contrast, a simple Markov model describing the transitions between but neglecting dynamical processes within conformational substates reproduced the observed distribution of first passage times. These findings suggest, that protein dynamics generally does not exhibit memory effects at time scales above a few hundred picoseconds, but confirms the existence of memory effects at a picosecond time scale.