AMBER Archive (2009)

Subject: Re: [AMBER] Langevin Dynamics in AMBER10

From: Carlos Simmerling (carlos.simmerling_at_gmail.com)
Date: Fri Jul 24 2009 - 07:01:28 CDT

it sounds like you're saying that you want to use a continuum solvent
instead of explicit. this isn't quite the same issue as deciding on a
thermostat.
there are a few things to keep in mind with continuum solvent MD:

- you almost certainly need to use the GB model, and although there are
multiple versions, all have various limitations that you should know about
before analyzing your data, so you know what might be an artifact of the
model as compared to "correct" behavior of your system
- read the literature to decide which of the GB models is best for your
system, and also pay attention to the choice of intrinsic radii
- the computation time per step of MD is probably slower in GB than with
explicit solvent. the only gain is that the solvent relaxes instantly, so
conformational changes that are limited by diffusion rates may be faster in
the continuum model. this is not the limiting behavior for many systems, so
make sure to think about what you want to model
- GB simulations often scale to more processors than explicit solvent
simulations, especially using pmemd
-until you become very comfortable with all of the issues, it's usually best
to choose peer-reviewed articles that successfully resolve problems that are
similar to what you want to do, and then use those as a guide for your
simulation protocols.

On Thu, Jul 23, 2009 at 7:12 PM, Biswaranjan Meher <brmeher_at_yahoo.co.in>wrote:

> Dear Dr. Simmerling,
> Thank you for your suggestion.
>
> In order to save time and the jobs to be a little faster compared to
> explicit solvation, I want to perform Langevin Dynamics simulation on a
> large protein without any water molecules . I am using an account in a
> cluster with 60# processors with Linux architecture of type x86_64 and
> total memory of 4052004 kB.
>
> Thank you again for your suggestion.
>
> regards
> Biswa
>
>
>
>
>
>
> --- On Fri, 24/7/09, Carlos Simmerling <carlos.simmerling_at_gmail.com>
> wrote:
>
> From: Carlos Simmerling <carlos.simmerling_at_gmail.com>
> Subject: Re: [AMBER] Langevin Dynamics in AMBER10
> To: "AMBER Mailing List" <amber_at_ambermd.org>
> Date: Friday, 24 July, 2009, 3:27 AM
>
> yes this will work fine. there are some possible things to think about, but
> we'd need to know more about what you want to do and the kind of computer
> you want to use.
>
> On Thu, Jul 23, 2009 at 5:54 PM, Biswaranjan Meher <brmeher_at_yahoo.co.in
> >wrote:
>
> > Dear AMBER Users,
> >
> > I want to perform Langevin Dynamics for my protein of interest containing
> > more than 1200 amino acid residues. Is it possible through AMBER10 ? if
> yes,
> > will it be a good choice to opt AMBER10 here ?
> >
> > Thanks for your suggestions
> >
> > regards
> > Biswa
> >
> >
> >
> >
> >
> > See the Web's breaking stories, chosen by people like you. Check out
> > Yahoo! Buzz. http://in.buzz.yahoo.com/
> > _______________________________________________
> > AMBER mailing list
> > AMBER_at_ambermd.org
> > http://lists.ambermd.org/mailman/listinfo/amber
> >
> _______________________________________________
> AMBER mailing list
> AMBER_at_ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
>
>
> Looking for local information? Find it on Yahoo! Local
> http://in.local.yahoo.com/
> _______________________________________________
> AMBER mailing list
> AMBER_at_ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
_______________________________________________
AMBER mailing list
AMBER_at_ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber