[gmx-developers] re: Energy and Foreign Lambdas and Hamiltonian Replica

[Berk Hess]

Hi,

That code is probably only for efficiently determining the potential 
energy at a different
lambda than that of the actual run. This is useful for determining the 
exchange probability.

But one could also just use the average dH/dlambda of the two replicas 
multiplied
by delta lambda. This is very easy to code.
I was actually thinking of doing this, but I currently don't need it myself.
To make it more efficient, you probably also want to be able to add a 
biasing potential
which is a function of lambda.

Berk.


servaas michielssens wrote:

> Thanks, for your fast reply and for the code.
> I had a look in the code but I can't find the swapping probability for 
> hamiltonian replica exchange. I expected this in repl_ex.c but I can't 
> find anything like it. I could not locate another function in md.c or 
> mdrun.c to do the hamiltonion replica exchange. Could you help me al 
> little further here were to look?
>  
>  
>  
> thanks!
>  
> servaas
>  
>
> Date: Fri, 15 Feb 2008 11:56:16 +0100
>
> Hi,
>
> My own server is unfortunately down since it was hacked last week and 
> we were anyway about to move from Joomla to Dekiwiki, but I've 
> uploaded a copy to the ftp site. Have a look in
>
> ftp://ftp.gromacs.org/pub/tmp/multilambda/
>
> Not sure if it's useful for your stuff, and since I haven't gotten 
> much feedback for it yet you shouldn't use it for production purposes 
> without significant testing. You've been warned, have fun ;-)
>
> Cheers,
>
> ERik
>
> Have a look at the
> On Feb 15, 2008, at 11:22 AM, servaas michielssens wrote:
>
> > Hi,
> >
> > I am planning to do a hamiltonian replica exchange simulation and I 
> > came acros this message. I am wondering what the status of this work 
> > is and if it is possible to obtain a copy of the code to do 
> > hamiltonain replica exchange. Or if not some suggestion on the best 
> > way to implement the extra energy evalution with a topoly on 
> > anonther node.
> >
> >
> > kind regards,
> >
> > servaas
> >
> >
> > Hi David,
> >
> > On Nov 2, 2007, at 5:19 PM, David Mobley wrote:
> > >
> > >> There is nearly no difference between the two as far as the replica
> > >> stuff is concerned.
> > >> It will be more work to (correctly) extract the foreign lambda 
> > energy
> > >> than it is to modify
> > >> the actual exchange criterion. Erik did some work on foreign lambda
> > >> calculations,
> > >> but I don't know what the status of that is.
> > >>
> > >> I guess the head branch would be the better option, especially if 
> > you
> > >> want to run
> > >> each replica over multiple cpu's.
> > >
> > > Maybe Erik can weigh in on whether this is "for sure", but my
> > > understanding is that he's almost done with the "foreign lambda"
> > > implementation (he indicated having a testable version over this
> > > coming weekend).
> >
> > Yes, I'll start to send this to David & John for testing, but in a
> > week or so we should have code for other people to play with too.
> >
> > Cheers,
> >
> > Erik
>
>This email was Anti Virus checked by Astaro Security Gateway. http://www.astaro.com
>
>------------------------------------------------------------------------
>
>_______________________________________________
>gmx-developers mailing list
>gmx-developers at gromacs.org
>http://www.gromacs.org/mailman/listinfo/gmx-developers
>Please don't post (un)subscribe requests to the list. Use the 
>www interface or send it to gmx-developers-request at gromacs.org.
>