There are several reasons for following that protocol. In my opinion, NPT
should always be used at some point during equilibration to allow the
density to settle into its *normal* value. This allows some leeway in
defining boxes and solvent packing that is not *ideal* for the starting
structure.
As for choosing NVT over NPT for production simulations: NVT has a direct
link to the helmholtz energy in statistical mechanics and corresponds to the
canonical partition function. It is a very natural ensemble. Practically
speaking (and it can be shown mathematically), as the ensembles become
infinitely large, the fluctuations become arbitrarily small and the mean is
always centered around the same point -- that is to say that all of the
ensembles are equivalent.
Also, NPT requires pressure coupling, which requires significantly more
computation per time step than just keeping the volume constant. Therefore,
since the answers won't change after your complex is sufficiently
equilibrated (especially true for condensed phase simulations since NVT ~
NPT for *nearly* non-compressible fluids), there is little benefit to the
increased cost of applying a barostat.
This is how I understand things, though others may certainly interject here.
Hope this helps,
Jason
On Sat, Jan 29, 2011 at 12:55 PM, Hirdesh Kumar <hirdesh.iitd.gmail.com>wrote:
> Thanks Sangita and Jason for your comments. Just another thing.. Recently I
> read a research paper in which the author did the NVT heating step,
> followed
> by long NPT equilibration and again NVT run before going to actual
> production. I am curious to know why one should proceed to NVT ensemble
> after NPT equilibration in any circumstances.
>
> Hirdesh
>
> On Sat, Jan 29, 2011 at 7:51 PM, case <case.biomaps.rutgers.edu> wrote:
>
> > On Sat, Jan 29, 2011, Hirdesh Kumar wrote:
> >
> > > I heated my system (protein of 329 residues) gradually over a period of
> > > ,000 steps using ntt=3 and gamma_ln = 1.0, with a weak restraint on the
> > > solute (i.e. 5 kcal/mol). Is there any need to remove the restraint
> > slowly
> > > using NVT ensemble further or I should directly proceed to NPT (
> without
> > any
> > > restraint on solute) ensemble as the restraint applied during heating
> was
> > > very slow.
> >
> > I would recommend keeing the restraints on during NPT. After you have a
> > good
> > density, I often run a further equilibration run with 0.5 kcal/mol-A^2
> > restraints, but that may just be habit...I haven't with/without tests.
> > After that you should run yet more equilibration with no restraints,
> before
> > starting an "production" runs. Don't skimp on equilibration: it can take
> > many, many nanoseconds for a big protein to adjust to the force field.
> >
> > ....dac
> >
> >
> > _______________________________________________
> > AMBER mailing list
> > AMBER.ambermd.org
> > http://lists.ambermd.org/mailman/listinfo/amber
> >
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
--
Jason M. Swails
Quantum Theory Project,
University of Florida
Ph.D. Graduate Student
352-392-4032
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Sat Jan 29 2011 - 12:30:02 PST
