[gmx-users] tabulated non-bonded potentials with replica exchange (potentials that change with temperature)

[Christopher Neale]

Dear Mark:

my mistake. I was using "-multidir 2" when really I should have been using "-multidir 0 1", which is why gromacs complained that I only had one replica.

Works as expected now.

Thank you again,
Chris.

________________________________________
From: Christopher Neale
Sent: 19 January 2015 17:45
To: Discussion list for GROMACS users
Subject: Re: [gmx-users] tabulated non-bonded potentials with replica exchange (potentials that change with temperature)

Dear Mark:

unfortunately is doesn't work in gmx4.6.7 (I didn't try 5.x). I can successfully run REMD with tables, though there seems no way to get different ensembles to pick up a different table each (see below). If you have any other ideas for how to do this, then that would be great.

# with -multi and tables in MD.xvg , MD_CL_CL.xvg :
-- works as expected, though each ensemble gets the same tables

# with -multi and tables in MD0.xvg , MD_CL_CL0.xvg , MD1.xvg , MD_CL_CL1.xvg :
Fatal error:
Library file MD.xvg not found in current dir nor in your GMXLIB path.

# with -multidir and no -multi :
Fatal error:
Need at least two replicas for replica exchange (option -multi)

# with -multi -multidir :
Fatal error:
mdrun -multi and -multidir options are mutually exclusive.

I've considered just commenting out the gmx_fatal call on line 621 of kernel/mdrun.c , but who knows what would happen then... probably something bad ;)

Thank you,
Chris.


________________________________________
From: gromacs.org_gmx-users-bounces at maillist.sys.kth.se <gromacs.org_gmx-users-bounces at maillist.sys.kth.se> on behalf of Mark Abraham <mark.j.abraham at gmail.com>
Sent: 19 January 2015 05:13
To: Discussion list for GROMACS users
Subject: Re: [gmx-users] tabulated non-bonded potentials with replica exchange (potentials that change with temperature)

On Mon, Jan 19, 2015 at 12:39 AM, Christopher Neale <
chris.neale at alum.utoronto.ca> wrote:

> Dear Users:
>
> I presume that I can use tabulated non-bonded potentials with replica
> exchange in gromacs 4.6.7.


Indeed, I can think of no reason why not (modulo that there's only support
for those in the group scheme).


> However, my question is whether it is possible to supply different tables
> for each of the different temperatures?
>

Can't think of a reason why or why not. I would be more hopeful of the
mdrun -multidir implementation (because that basically does a "cd" at the
start of mdrun, so it doesn't need all the table filenames to have been
suffixed correctly, they just need you to organize them correctly).


> My actual use involves the plumed plugin to do hamiltonian replica
> exchange. With that patch it is possible to modify the .tpr files for each
> ensemble prionr to calling mdrun and have gromacs mdrun compute exchange
> energies properly. However, since tables are sent to mdrun directly from
> the command line, rather than through the .tpr file, my guess is that there
> is no way to load in a separate tabulated potential for each ensemble.
>

Embedding table files in the .tpr is on the to-do list for implementing
tables in the Verlet scheme.

Mark


>
> Thank you for your insight,
> Chris.
> --
> Gromacs Users mailing list
>
> * Please search the archive at
> http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before
> posting!
>
> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>
> * For (un)subscribe requests visit
> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or
> send a mail to gmx-users-request at gromacs.org.
>
--
Gromacs Users mailing list

* Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting!

* Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

* For (un)subscribe requests visit
https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-request at gromacs.org.