[gmx-developers] The role of Jacobian factor in alchemical free energy simualtion

[Berk Hess]

Hi,

If this is an "issue" or not depends on what you want to calculate.

I have never encountered users that want to change many constraints 
lengths in a molecule and also we would in most cases advise people to 
only constrain bonds involving hydrogens, so I don't think this is an 
important issue.

Cheers,

Berk

On 09/15/2016 04:10 PM, BOSISIO Stefano wrote:
>
> Thanks for the answer.
>
> I am still wondering 2 questions:
>
> 1) Calculating the Fixman potential could lead to a correction of this 
> issue?
>
> 2) Is Gromacs programmed to correct this issue?
>
> Thank you very much
>
>
> Best,
>
>
> Stefano
>
> ------------------------------------------------------------------------
> *From:* gromacs.org_gmx-developers-bounces at maillist.sys.kth.se 
> <gromacs.org_gmx-developers-bounces at maillist.sys.kth.se> on behalf of 
> Berk Hess <hess at kth.se>
> *Sent:* 13 September 2016 16:54:39
> *To:* Discussion list for GROMACS development
> *Subject:* Re: [gmx-developers] The role of Jacobian factor in 
> alchemical free energy simualtion
>
> No, all other interactions that work on the bond length also 
> contribute, i.e. forces from potentials and other connected constraints.
>
> Cheers,
>
> Berk
>
>
> On Sep 13, 2016 5:24 PM, BOSISIO Stefano <s1374284 at sms.ed.ac.uk> wrote:
>
>     Hi,
>
>
>     Thanks for the answer.
>
>     Thus, taking a constrained bond length, which is alchemically
>     transformed, the only term that comes into play is the mass metric
>     tensor, isn't it?
>     Is there any piece of code of Gromacs which is designated to
>     correct this issue, you could address me?
>
>
>
>     Best wishes,
>
>
>     Stefano
>
>     ------------------------------------------------------------------------
>     *From:* gromacs.org_gmx-developers-bounces at maillist.sys.kth.se
>     <gromacs.org_gmx-developers-bounces at maillist.sys.kth.se> on behalf
>     of Berk Hess <hess at kth.se>
>     *Sent:* 13 September 2016 12:02
>     *To:* gmx-developers at gromacs.org
>     *Subject:* Re: [gmx-developers] The role of Jacobian factor in
>     alchemical free energy simualtion
>     Hi,
>
>     In alchemical reactions it is often desirable to avoid changing
>     bond lengths by using a topology with atoms from both states
>     present. In that case there is no explicit Jacobian involved. If
>     you do change constraint lengths, the metric tensor contribution
>     comes in through the constraint forces. If you want this effect or
>     not depends on what quantity you want to calculate exactly. You
>     can correct for this analytically afterwards.
>
>     Cheers,
>
>     Berk
>
>     On 2016-09-13 11:57, BOSISIO Stefano wrote:
>
>         Dear Gromacs staff,
>
>         I am trying to understand some internal issues between my code
>         and Gromacs in alchemical free energy calculations.
>
>         Considering a simple alchemical free energy calculation (e.g
>         ethane to methanol) with a constraint applied to all the
>         hydrogen bonds  does Gromacs calculate a Jacobian correction ?
>         (Boresch Stefan, and Martin Karplus "The Jacobian factor in
>         free energy simulations." /The Journal of chemical
>         physics/105.12 (1996): 5145-5154  )
>         Alternatively, does Gromacs apply any correction to the
>         computed free energy changes if constraints were applied to a
>         solute?
>
>
>         Thank you
>
>         Best regards,
>
>         Stefano
>
>
>
>         The University of Edinburgh is a charitable body, registered in
>         Scotland, with registration number SC005336.
>
>
>
>
>
>
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-developers/attachments/20160915/67725887/attachment-0001.html>