On Mon, Aug 13, 2012 at 4:42 PM, George Tzotzos <gtzotzos.me.com> wrote:
> Hi everybody,
>
> I know that there is no prescribed answer to my question but I'd still
> like to ask as my computational power is rather limited (running a 24 core
> Mac)
>
> In calculating free energies of binding (MMPBSA) is it preferable to run a
> long trajectory or several smaller ones? I'm dealing with a proteins of
> ~150 residues.
>
Genheden et. al has discussed this point (DOI 10.1002/jcc.21366) and come
to the conclusion that several short simulations give rise to more
statistically converged answers, since you obtain more uncorrelated
structures that way.
Of course if you started with the same structure and ran only 10 fs, you
would get statistically converged answers as well (in that the standard
deviation would be small), since they're all effectively the same structure
(and you never know, the various snapshots may even appear uncorrelated :)
). That's obviously not what they did there (they ran ~200 ps per
simulation, IIRC), but I'm still not entirely convinced.
HTH,
Jason
> Your experience is valued
>
> George
>
>
>
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
--
Jason M. Swails
Quantum Theory Project,
University of Florida
Ph.D. Candidate
352-392-4032
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon Aug 13 2012 - 19:30:04 PDT
