[gmx-developers] Re: [gmx-users] 'the reciprocal space calculation will not necessarily cancel'

Floris Buelens floris_buelens at yahoo.com
Fri May 17 13:03:27 CEST 2013


Thanks, that helps a lot. If I understand correctly, everything is fine at the end points (lambda 0 and 1), while at intermediate states the soft core electrostatics close to the cutoff can be marginally off. However the potential is still well-defined and contributions to dU/dlambda are correctly summed, so that in practice we're talking about a tiny deviation from the regular soft core transformation pathway, to another one which is still thermodynamically rigorous albeit with a very slight negative effect on energy conservation. Does that sound about right?




----- Original Message -----
From: Justin Lemkul <jalemkul at vt.edu>
To: Floris Buelens <floris_buelens at yahoo.com>; Discussion list for GROMACS development <gmx-developers at gromacs.org>
Cc: 
Sent: Thursday, 16 May 2013, 14:05
Subject: Re: [gmx-developers] Re: [gmx-users] 'the reciprocal space calculation
 will not necessarily	cancel'



On 5/16/13 7:54 AM, Floris Buelens wrote:
> Hi,
>
> I received no reply on the users list, could anybody enlighten me on the warning message in the email below?
> thanks!
>

The warning was made more clear before version 4.6.1; I assume you are using 
4.6?  There is discussion of what is going on at 
http://redmine.gromacs.org/issues/1146.  The change Berk implemented created a 
totally new warning message with more explicit information:

http://redmine.gromacs.org/projects/gromacs/repository/revisions/a46b7f1f2f380c7238fe63bce706078343c9210e/diff/src/kernel/readir.c

-Justin

>
> ----- Original Message -----
> From: Floris Buelens <floris_buelens at yahoo.com>
> To: "gmx-users at gromacs.org" <gmx-users at gromacs.org>
> Cc:
> Sent: Tuesday, 14 May 2013, 14:02
> Subject: [gmx-users] 'the reciprocal space calculation will not necessarily
>     cancel'
>
> Hi,
>
> I'm not clear on the meaning and implications of this warning message from grompp:
>
> WARNING 1 [file ../md.mdp]:
>
>    With coulomb soft core, the reciprocal space calculation will not
>    necessarily cancel.  It may be necessary to decrease the reciprocal space
>    energy, and increase the cutoff radius to get sufficiently close matches
>    to energies with free energy turned off.
>
> Can anybody help me understand this better? In what context are we talking about cancellation? Around a thermodynamic cycle, or in a more narrow sense? Why would we not see such a cancellation?
>
> In my understanding, if we have an A state with a certain set of charges and a B state with another, the end states at lambda=0 and lambda=1 should match exactly what you would see if you evaluated the A and B states independently without using the free energy code. Soft core vdW or coulomb would only affect the shape of the pathway connecting them and not the end points. So I'm having trouble seeing what kind of cancellation we're relying upon.
>
> Thanks in advance for any insights!
> --
> gmx-users mailing list    gmx-users at gromacs.org
> http://lists.gromacs.org/mailman/listinfo/gmx-users
> * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
> * Please don't post (un)subscribe requests to the list. Use the
> www interface or send it to gmx-users-request at gromacs.org.
> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>

-- 
========================================

Justin A. Lemkul, Ph.D.
Research Scientist
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu | (540) 231-9080
http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin

========================================




More information about the gromacs.org_gmx-developers mailing list