On Mon, Sep 13, 2010, M. Reza Ganjalikhany wrote:
>
> I want to study the effect of different temperatures on structure and
> flexibility of a protein containing 270 aa by MD simulation in explicit
> solvent. Since I'm investigating the effect of temperature, I found it
> difficult to chose the best way among several methods in heating and
> equilibration which are presented in papers and tutorials.
>
> 1-Is it better to start heating from 0K or higher temperatures like 100K or
> 200K? (to prevent wild fluctuations and to be well equilibrated)
There is no correct answer here; it probably makes little difference which you
choose. Be sure to equilibrate fully at the final temperature of interest.
> 2-Is it better to heat up the system directly (from beginning to final
> temperature) or do it in several steps?
Same answer.
> 3-Is it better to heat up the system gradually (in a controlled manner
> nmropt=1) or as routine procedure?
Ditto.
> 4- how long should be sufficient for each step for a protein like this?
Probably a long time, since you want reliable (small) differences between
different temperatures. You will probably have to generate your own
statistics, running multiple simulations (say with different heating
profiles), then running them long enough at the final temperature to show that
things no longer depend on the initial heating step, and that differences
between temperatures are reprodicible.
....dac
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon Sep 13 2010 - 11:30:03 PDT
