My 2 cents on NpT:
It seems to be considered AMBER best practice to do *some* NVT before doing
any NpT. Presumably this is a numerical stability thing under the
assumption that low Boltzmann weight conformations can lead to nasty volume
fluctuations. However, the opposite is also true; low density water boxes
(like those produced by LEAP) can lead to vacuum bubbles if you run NVT for
too long.
The use of the term "NpT" should be used with caution. <Puts on physicist
hat> The only available pressure algorithm in AMBER is the Berendsen
barostat, which, as far as I know, has never actually been claimed to
produce configurations with the proper Boltzmann weight. In my opinion it
is more proper to say (and Berendsen uses language similar to this in the
original paper) that the algorithm performs "weak coupling to a heat bath
at constant pressure." <Takes off physicist hat>
That being said, I've made comparisons to reported values from other
algorithms like Andersen's and average quantities (like density and
potential energy) of water boxes come out pretty much exactly the same.
The fluctuations, however, are probably wrong; so don't go trying to report
isobaric heat capacities or isothermal compressibilities or something else
crazy. Since AMBER is generally not used for those things, it's usually a
"no harm no foul" scenario to use the term NpT.
Regards,
Brian
PS. Those 2 cents were in Canadian.
On Sat, Feb 18, 2012 at 10:47 AM, Daniel Roe <daniel.r.roe.gmail.com> wrote:
> On Sat, Feb 18, 2012 at 9:11 AM, Aron Broom <broomsday.gmail.com> wrote:
> > I can't really imagine that the default would
> > be 0 as that must mean a number of people think they are running langevin
> > dynamics and are not, but that was my impression from the manual.
>
> The default value for gamma_ln actually is 0. The 'ntt = 3' part of
> the manual does call attention to this:
>
> "Note that when <gamma> has its default value of zero, this is the
> same as setting ntt = 0"
>
> I suppose that when ntt=3 the code could set gamma_ln to something
> automatically, but I think the reasoning behind leaving it at 0 is
> that the 'best' value for gamma_ln changes depending on what you want
> your thermostat to do. For GB simulations where you just want to add
> some random friction to prevent "hotspots" while still having
> relatively fast conformational sampling, a value of around 5 is
> probably fine. If you need a collision frequency approaching that of
> water you'll need something around 50-80. Of course, this also varies
> depending on who you ask.
>
> -Dan
>
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
--
================================ Current Address =======================
Brian Radak : BioMaPS
Institute for Quantitative Biology
PhD candidate - York Research Group : Rutgers, The State
University of New Jersey
University of Minnesota - Twin Cities : Center for Integrative
Proteomics Room 308
Graduate Program in Chemical Physics : 174 Frelinghuysen Road,
Department of Chemistry : Piscataway, NJ
08854-8066
radak004.umn.edu :
radakb.biomaps.rutgers.edu
====================================================================
Sorry for the multiple e-mail addresses, just use the institute appropriate
address.
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon Feb 20 2012 - 07:00:02 PST
