[gmx-users] energygrps interactions are 0 for water

[Mark Abraham]

Hi,

You're probably running on GPUs, and have missed the note in the .log file
that observes that energy groups are not supported there (it'd be too slow
to run).

If you want such data, then the recommend procedure is to run your
simulation normally, then use gmx mdrun -rerun old-trajectory -nb cpu,
which does the energy decomposition you want.

Mark

On Thu, Sep 17, 2015 at 7:00 PM Julian Michalowsky <j.michalowsky at gmail.com>
wrote:

> Hi,
>
> I have a problem regarding the calculation of interaction energies using
> gmx energy and the energygrps parameter in the mdp file. My mdp file
> includes the line
>
> energygrps = Ion OW
>
> Whenever I calculate the LJ-SR and Coul.-SR energy terms between OW
> and any other energy group, the result is just zero. I am using an
> appropriate
> index file during preprocessing that defines OW as a group and also tried
> SOL and Water as energygrps in the mdp file.
>
> I originally used this with a custom force field, but encountered the same
> problem during a test run using the gromos54a7.ff from the default top
> directory.
>
> The system is a simple electrolyte solution (NaCl in water). I use a python
> script to set up the initial gro file. Ions are referred to as NA, CL in
> both gro
> and top file (not NA+ or CL-, if that makes any difference). The top file
> includes the respective forcefield.itp, ions.itp and spce.itp topologies
> via
> the #include statement using absolute paths. Gromacs version is 5.0.
>
> I also tried inserting the ions into a box with water using gmx genion. The
> results are the same: Ion-Ion interactions are present and non-zero, but
> Ion-SOL, SOL-SOL etc. are zero. The problem arises using various
> temperature
> and pressure coupling algorithms. Apart from that, the simulations seem to
> run smoothly and not doing exceptionally weird things, even reproduce
> results.
>
> I am unable to grasp the source of the issue here, so if you can help, any
> input is welcome. Presumably it's just a mistake regarding input on my
> part,
> but I have been trying to solve this problem for quite some time now, so
> I'm
> writing to the mailing list.
>
> Thanks in advance and kind regards,
> Julian Michalowsky
> --
> 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.
>