[gmx-developers] Interface - energy value from peptide conformation

[Martin Kamp Jensen]

Hi Mark,

Thanks for your answer. I was not being clear in my first mail. Please see
below.

On Fri, Sep 24, 2010 at 10:37 AM, Mark Abraham <mark.abraham at anu.edu.au>wrote:

>
>
> ----- Original Message -----
> From: Martin Kamp Jensen <martin.kamp.jensen at gmail.com>
> Date: Friday, September 24, 2010 18:23
> Subject: [gmx-developers] Interface - energy value from peptide
> conformation
> To: gmx-developers at gromacs.org
>
> > Hello,
>
> > I just started experimenting with GROMACS and I would like to interface
> with it to get energy values for different conformations of peptides. When
> debugging my way through the energy minimization part of the gmxdemo
> tutorial, I ended up in the following method of the mdlib/minimize.c file:
> >
> > static void evaluate_energy(FILE *fplog,bool bVerbose,t_commrec *cr,
> >     t_state *state_global,gmx_mtop_t *top_global,
> >     em_state_t *ems,rvec **buf,gmx_localtop_t *top,
>  >     t_inputrec *inputrec,
> >     t_nrnb *nrnb,gmx_wallcycle_t wcycle,
> >     gmx_vsite_t *vsite,gmx_constr_t constr,
> >     t_fcdata *fcd,
> >     t_graph *graph,t_mdatoms *mdatoms,
> >     t_forcerec *fr, rvec mu_tot,
> >     gmx_enerdata_t *enerd,tensor vir,tensor pres,
> >     int count,bool bFirst)
> >
> > I'm not very experienced with C/C++, but the fact that this method is
> static means that I cannot just call it from another C program (or e.g. a
> Java program) without changing GROMACS. I guess a naive way would be setting
> up an energy minimization and just running one step, but that is a lot of
> overhead and not very clean. Do you have any pointers for me?
> >
> > (Actually, what I want to do is navigate the energy landscape of
> peptides by changing the torsion angles. This means that I will need to look
> up energy values very frequently.)
>
> The elegant way to do that is to write your series of conformations to a
> pseudo-trajectory (in whichever of GROMACS many trajectory formats is
> convenient to you), and then invoke mdrun -rerun on that trajectory with a
> suitable .tpr. Then you either parse the .log file or the output .edr file
> for the energies. This should keep the overhead quite low.
>

This seems like a good idea when you have all the conformations for which
you want to look up energy values. The thing is, I need to get an energy
value, then do some work outside of GROMACS, get a new energy value, etc.

Imagine you have two conformations in the energy landscape, both are local
minima. Then I would like to find out how a conformation should change to
get from one to another while trying to ensure low energy values for the
intermediate conformations. This will require the use of heuristics that
will need to look up the energy value in each step where a conformation will
be changed slightly from one towards another.


> Mark
>
> --
> gmx-developers mailing list
> gmx-developers at gromacs.org
> http://lists.gromacs.org/mailman/listinfo/gmx-developers
> Please don't post (un)subscribe requests to the list. Use the
> www interface or send it to gmx-developers-request at gromacs.org.
>

Regards,
Martin.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-developers/attachments/20100924/24aa3ca8/attachment.html>