Re: how long should the simulation be run

From: Peter Freddolino (petefred_at_ks.uiuc.edu)
Date: Sat Apr 29 2006 - 16:22:11 CDT

Hi Santhu,
this is a very difficult question that is greatly dependent on what
you're trying to do. In general there are 3 phases to any of these
simulations: minimization, equilibration, and dynamics. Minimization
should be run until it converges to some fairly low rms value, and
equilibration should be run until initial instabilities work themselves
out, ie, diffusive molecules start following the diffusion law, and
protein rmsd stabilizes from any initial changes. This usually takes on
the order of tens of picoseconds. After that you get into your
production run, which is going to vary from one to hundreds or thousands
of nanoseconds depending on what you want to see. I'd highly recommend
consulting a book like Leach's _Molecular Modeling: Principles and
Applications_ for more detail on what's going on in all of these stages.

If you saw a significant change in 6 ps, it would tend to indicate that
your starting state wasn't particularly stable (which may be the point).

Best,
Peter

genie lfs wrote:

> hello ..
> i am new to simulations and have tried a few simulations to understand.
> i understand the parameters involved and how to run a simulation.
>
> but how long should a simulation be run ? like for a energy
> minimization once we get a stable energy we can safely say that the
> system is minimized..
>
> i am using simulations to understand protein-rna interactions .. i ran
> the simulation for 10 ps and didnot find anything unusal after the 6
> ps .. what ever was expected to happen according to theory happened in
> 6 ps and the rsst 4 ps the simulation was of no interest.
>
> but most of the simulations i see run for 1ns .. so can you please
> guide me over this.
>
> thanking you
> santhu

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:43:34 CST