[gmx-developers] Zeros in interaction enegy calculations

[Guy Vigers]

Filed as issue #4462.

Thank you very much!

Guy Vigers


> On Apr 6, 2022, at 7:27 AM, Berk Hess <hess at kth.se> wrote:
> 
> Hi,
> 
> I just tried an both a standard mdrun and a rerun produce correct group pair energies for me in versions 2020, 2021 and 2022.
> 
> Molecules have nothing to do with this. The energy groups are atom indices, it does not matter where they come from.
> 
> Please file an issue on gitlab and attach all files required to run grompp.
> 
> Cheers,
> 
> Berk
> 
> On 2022-04-06 15:10 , Guy Vigers wrote:
>> Dear Gromacs Developers
>> 
>> There are at least 3 threads on the user forum where people attempt to calculate interaction energies on a subset of their system and get back all zeros.
>> 
>> https://gromacs.bioexcel.eu/t/zero-interaction-energy-from-mdrun-rerun-in-gmx-2020-vs-non-zero-energies-in-gmx-2018/2396/6
>> 
>> https://gromacs.bioexcel.eu/t/energy-calculations-of-residues-is-zero/2474
>> 
>> https://gromacs.bioexcel.eu/t/how-to-get-potential-energy-of-a-single-water-molecule-with-the-rest-of-the-system/1492
>> 
>> 
>> I'm seeing the same thing in my work. The usual explanation is that you can't do interaction energy calculations on a GPU, but neither I nor these other users appear to have made that mistake. There is some indication that it might be related to multiple molecules with the same name (but different atom indices, of course) but I'm not sure of that.
>> 
>> According to one of the threads, the calculation works correctly in Gromacs 2018 but not in 2020. I have not verified that, but I’m getting all zeros in 2021 and 2022.
>> 
>> I'd like to try and help troubleshoot this problem. I can supply all of the relevant files to reproduce it. Is there an expert who would be willing to work on this with me?
>> 
>> BTW Thank you for the great program!
>> 
>> Best Wishes
>> 
>> Guy Vigers
>> 
> 
> -- 
> Gromacs Developers mailing list
> 
> * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-developers_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-developers or send a mail to gmx-developers-request at gromacs.org.