On Tue, Mar 15, 2016, anu chandra wrote:
>
> I am using the following input parameters for doing the production run of
> classical MD simulation with Amber.
>
> imin = 0, ntx = 5, irest = 1,
> ntt =3, gamma_ln=2.0, ig=-1,
>
> Following a standard protocol, I did minimization and extensive (10-20 ns)
> equilibration of the protein-water system before entering to production
> run. But, I am a little confused about the use random seed generator in
> production. As per Amber manual, ig=-1 will generate different random seed
> and thereby different staring velocity for each simulation window ( say one
> use a 5 ns window for doing a 100 ns long simulation). Such a change in
> velocity can bring changes in the phase space for every run ( ie, the run
> may not be a continuation of previous window). If that the case, what is
> the point of doing extensive equilibration here?
When you set irest=1, the initial velocities for that run are taken from the
restart file, and no random number is used. For you case, the random number
generator is only used for the Langevin integrator.
...hope this helps....dac
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Tue Mar 15 2016 - 07:00:03 PDT
