[gmx-developers] Andersen Barostat

Thu Nov 26 15:57:09 CET 2009

Hi, Stefan-

---------------------------------------------
We are involved in a project aiming where we need to employ an
Anderson barostat and to use the Bennett Acceptance Ratio method to
compute free energies using GROMACS. From the GROMACS wiki and mailing
list it appears that you have put work into some areas related to
these areas and have already written some of the code that would be
useful to us. Would it be possible for you to give us access to the
code if it is in in a usable state? We would be willing to contribute
to the development and testing if it were.
-----------------------------------------------

Parinello-Rahman is an extension of the Anderson thermostat to
arbitrary stress/strain matrices, instead of just isotropic pressures.
 Pressure control currently in Gromacs has one issue, in that the
pressure is calculated from the last step.

I have an (alpha) version of a velocity verlet implementation of the
Martyna-Tuckerman-Tobias-Klein version of the Anderson Thermostat.  If
you can wait a week or two, it will likely be in the main CVS by then;
if you not you can access it at:

git clone git://git.gromacs.org/~mrshirts/gromacs_mrs.git

Note that the parallel version does not quite work yet! If you want to
help test it before it's added to the CVS, then I am happy to help
out.

For Bennett Acceptance Ratio, all you need is the energies evaluated
the other lambda states to be output.  This is supported in both the
CVS version of gromacs and my branch.  Just add the keyword
"foreign-lambdas" and add the other lambda values you want to evaluate
the energy at.

To analyze these, you can use the pymbar tools found at
https://simtk.org/home/pymbar; look at the alchemical free energy
example.  It will have to be modified a little bit to match the
current gromacs code, however!

This request should be much easier in a short time, but not quite yet . . .
--

Stefan Zasada <stefan.zasada at ucl.ac.uk>
Centre for Computational Science
Chemistry Department, University College London,
20 Gordon Street, London, WC1H 0AJ, UK.
Tel: +44 (0)20 7679 5300

On Thu, Nov 26, 2009 at 6:00 AM,  <gmx-developers-request at gromacs.org> wrote:
> Send gmx-developers mailing list submissions to
>        gmx-developers at gromacs.org
>
> To subscribe or unsubscribe via the World Wide Web, visit
>        http://lists.gromacs.org/mailman/listinfo/gmx-developers
> or, via email, send a message with subject or body 'help' to
>        gmx-developers-request at gromacs.org
>
> You can reach the person managing the list at
>        gmx-developers-owner at gromacs.org
>
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of gmx-developers digest..."
>
>
> Today's Topics:
>
>   1. Gromacs/gaussian interface (Ake Sandgren)
>   2. Re: Gromacs/gaussian interface (Gerrit Groenhof)
>   3. Andersen Barostat (Stefan Zasada)
>   4. pV term calculated incorrectly (Michael Shirts)
>   5. Re: pV term calculated incorrectly (David van der Spoel)
>   6. gromacs/gaussian09 (Ake Sandgren)
>
>
> ----------------------------------------------------------------------
>
> Message: 1
> Date: Wed, 25 Nov 2009 15:40:48 +0100
> From: Ake Sandgren <ake.sandgren at hpc2n.umu.se>
> Subject: [gmx-developers] Gromacs/gaussian interface
> To: gmx-developers at gromacs.org
> Message-ID: <1259160048.1773.1377.camel at skalman.hpc2n.umu.se>
> Content-Type: text/plain
>
> Hi!
>
> I know this is mostly out of scope of this list but...
>
> Does anyone have a working gromacs/gaussian (03 or 09) setup which they
> could run a short qmmm test on?
>
> --
> Ake Sandgren, HPC2N, Umea University, S-90187 Umea, Sweden
> Internet: ake at hpc2n.umu.se   Phone: +46 90 7866134 Fax: +46 90 7866126
> Mobile: +46 70 7716134 WWW: http://www.hpc2n.umu.se
>
>
>
> ------------------------------
>
> Message: 2
> Date: Wed, 25 Nov 2009 15:55:49 +0100
> From: Gerrit Groenhof <ggroenh at gwdg.de>
> Subject: Re: [gmx-developers] Gromacs/gaussian interface
> To: Discussion list for GROMACS development
>        <gmx-developers at gromacs.org>
> Message-ID: <4B0D4575.5010104 at gwdg.de>
> Content-Type: text/plain; charset=ISO-8859-1
>
> Sure, sned me the tpr.
>
> Gerrit
>
> Ake Sandgren wrote:
>> Hi!
>>
>> I know this is mostly out of scope of this list but...
>>
>> Does anyone have a working gromacs/gaussian (03 or 09) setup which they
>> could run a short qmmm test on?
>>
>>
>
>
>
> ------------------------------
>
> Message: 3
> Date: Wed, 25 Nov 2009 16:50:58 +0000
> From: Stefan Zasada <stefan.zasada at ucl.ac.uk>
> Subject: [gmx-developers] Andersen Barostat
> To: gmx-developers at gromacs.org
> Cc: David Wright <dave.william.wright at googlemail.com>
> Message-ID: <7B6FE686-3C05-47A7-809C-811A09128478 at ucl.ac.uk>
> Content-Type: text/plain; charset=us-ascii; format=flowed; delsp=yes
>
> Dear GROMACS developers,
>
> We are involved in a project aiming where we need to employ an
> Anderson barostat and to use the Bennett Acceptance Ratio method to
> compute free energies using GROMACS. From the GROMACS wiki and mailing
> list it appears that you have put work into some areas related to
> these areas and have already written some of the code that would be
> useful to us. Would it be possible for you to give us access to the
> code if it is in in a usable state? We would be willing to contribute
> to the development and testing if it were.
>
> Thanks,
>
>
>
> Stefan.
>
> --
> Stefan Zasada <stefan.zasada at ucl.ac.uk>
> Centre for Computational Science
> Chemistry Department, University College London,
> 20 Gordon Street, London, WC1H 0AJ, UK.
> Tel: +44 (0)20 7679 5300
>
>
>
>
>
> ------------------------------
>
> Message: 4
> Date: Wed, 25 Nov 2009 22:11:28 -0500
> From: Michael Shirts <michael.shirts at virginia.edu>
> Subject: [gmx-developers] pV term calculated incorrectly
> To: Discussion list for GROMACS development
>        <gmx-developers at gromacs.org>
> Message-ID:
>        <662e7a930911251911t75dce826vb294ad69828905ea at mail.gmail.com>
> Content-Type: text/plain; charset=ISO-8859-1
>
> I noticed that the pV term stored in the mdebin when a NPT simulation
> is performed is calculated incorrectly -- it uses the instantaneous
> pressure, when it should be using the applied pressure.  dU = pdV, so
> with constant p, U = \int p dV = p \int dV = pV
>
> I've written a fix (with full generality for arbitrary stress/strain),
> but have realized I'm not exactly sure which branches I should be
> checking into, and can't figure out exactly which git command I should
> be using from the gromacs web page.  Any hints?
>
> Best,
> Michael
>
>
> ------------------------------
>
> Message: 5
> Date: Thu, 26 Nov 2009 08:33:43 +0100
> From: David van der Spoel <spoel at xray.bmc.uu.se>
> Subject: Re: [gmx-developers] pV term calculated incorrectly
> To: Discussion list for GROMACS development
>        <gmx-developers at gromacs.org>
> Message-ID: <4B0E2F57.6020106 at xray.bmc.uu.se>
> Content-Type: text/plain; charset=ISO-8859-1; format=flowed
>
> Michael Shirts wrote:
>> I noticed that the pV term stored in the mdebin when a NPT simulation
>> is performed is calculated incorrectly -- it uses the instantaneous
>> pressure, when it should be using the applied pressure.  dU = pdV, so
>> with constant p, U = \int p dV = p \int dV = pV
>>
>> I've written a fix (with full generality for arbitrary stress/strain),
>> but have realized I'm not exactly sure which branches I should be
>> checking into, and can't figure out exactly which git command I should
>> be using from the gromacs web page.  Any hints?
>>
>> Best,
>> Michael
> I'm not sure that anyone is using it, but thanks for fixing. In
> principle it is sufficient if it goes in the master branch I would think.
>
> --
> David van der Spoel, Ph.D., Professor of Biology
> Molec. Biophys. group, Dept. of Cell & Molec. Biol., Uppsala University.
> Box 596, 75124 Uppsala, Sweden. Phone:  +46184714205. Fax: +4618511755.
> spoel at xray.bmc.uu.se    spoel at gromacs.org   http://folding.bmc.uu.se
>
>
> ------------------------------
>
> Message: 6
> Date: Thu, 26 Nov 2009 11:34:24 +0100
> From: Ake Sandgren <ake.sandgren at hpc2n.umu.se>
> Subject: [gmx-developers] gromacs/gaussian09
> To: gmx-developers at gromacs.org
> Message-ID: <1259231664.1773.1431.camel at skalman.hpc2n.umu.se>
> Content-Type: text/plain
>
> Hi!
>
> The instructions for gromacs/gaussian says gaussian03 (at least the only
> ones i've found so far)
> Are there new instructions for gaussian09 somewhere?
>
> --
> Ake Sandgren, HPC2N, Umea University, S-90187 Umea, Sweden
> Internet: ake at hpc2n.umu.se   Phone: +46 90 7866134 Fax: +46 90 7866126
> Mobile: +46 70 7716134 WWW: http://www.hpc2n.umu.se
>
>
>
> ------------------------------
>
> --
> gmx-developers mailing list
> gmx-developers at gromacs.org
> http://lists.gromacs.org/mailman/listinfo/gmx-developers
>
>
> End of gmx-developers Digest, Vol 67, Issue 20
> **********************************************
>