[gmx-users] g_energy menu choices inconsistent?

[Justin Lemkul]

On 10/2/12 1:40 AM, Ladasky wrote:
> I have been trying to automate my simulation setup and monitoring.  I wrote a
> script which calls g_energy, and automatically generates plots of potential
> energy from my EM step, temperature from my NVT equilibration step, and
> pressure and density from my NPT equilibration step.  For each of these
> three simulations, it appeared that the numbering on the g_energy menu was
> consistent:
>
> Potential = 11, Temperature = 15, Pressure = 16, Density = 22.
>
> I wrote my program assuming that these numbers would always describe the
> parameters listed.  Maybe not?
>
> Something is still going wrong with my production run, I'm getting a
> segmentation fault after about 40,000 steps (sigh, I almost thought I had
> this whole setup understood and debugged).  So I used my graphing program to
> check the temperature, and the results were... odd, bouncing from +300 to
> -100.  The values in my MD.log file disagreed with my graph, they were all
> within a few degrees of 310K, my target temperature.
>
> So I re-ran g_energy directly, bypassing my program, using my production-run
> ener.edr file as input.  The menu showed:
>
> Potential = 10, Temperature = 13, Pressure = 15, Density = 21.
>
> Seeing this number shift, I understand why my "temperature" graph for the
> production looked wrong -- I was actually graphing pressure.
>
> Not knowing what menu options to expect will impair my ability to automate
> the running of g_energy.  I can try some tricks to capture that menu each
> time I run the program, but it will be a chore.
>
> I would also like to understand: what is the REASON that these numbers might
> change between runs?
>

Not all energy terms are written if they are not applicable.  This saves disk 
space by making the .edr file smaller if it can be.

Note that you can always select by name rather than number, i.e.:

echo Temperature | g_energy -f ener.edr

-Justin

-- 
========================================

Justin A. Lemkul, Ph.D.
Research Scientist
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu | (540) 231-9080
http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin

========================================