Re: [AMBER] MMPBSA.py.MPI hangs

From: George Tzotzos <gtzotzos.me.com>
Date: Tue, 25 Feb 2014 19:16:30 -0300

Jason, Bill,

Thanks a lot. I have made several repetitions. It always hangs.

In the last couple, it aborted on rank 10. I’m attaching the _MMPBSA_complex_gb.mdout.10 and 11.
I’m also attaching the input script. In the last run, I set mol surf=1
Finally, I’m also attaching the prmtop and inpcrd files.

Once again thanks for looking into the problem.

Regards

George




On Feb 25, 2014, at 5:21 PM, Jason Swails <jason.swails.gmail.com> wrote:

> On Tue, 2014-02-25 at 13:52 -0300, George Tzotzos wrote:
>> Hi everybody,
>>
>> I checked the archive to see if I could identify the source of error occurring “on rank 0”. I didn’t find anything that really makes sense to me. The prmtop file identified in the log (see below) loads correctly in VMD and Chimera.
>>
>> Any help would be much appreciated.
>>
>> Last but not least all patches have been applied successfully.
>>
>> Log file output follows below
>>
>> Thanks in advance
>>
>> George
>>
>>
>> Loading and checking parameter files for compatibility...
>> mmpbsa_py_energy found! Using /Users/3i521409/Programs/amber12/bin/mmpbsa_py_energy
>> cpptraj found! Using /Users/3i521409/Programs/amber12/bin/cpptraj
>> Preparing trajectories for simulation...
>> 1000 frames were processed by cpptraj for use in calculation.
>>
>> Running calculations on normal system...
>>
>> Beginning GB calculations with /Users/3i521409/Programs/amber12/bin/mmpbsa_py_energy
>> calculating complex contribution...
>> CalcError: /Users/3i521409/Programs/amber12/bin/cpptraj failed with prmtop 4fqt_OVT.prmtop!
>
> It seems like something went wrong with the LCPO calculation. Try
> setting molsurf=1 in the &gb section of your input file. Alternatively
> if the problem was with molsurf try using LCPO instead. (The surface
> area calculation is run with cpptraj as of AmberTools 13).
>
> In the meantime, can you send me the 4fqt_OVT.prmtop file and a
> structure for it? (A restart, PDB, or trajectory with 1 frame works).
>
> Thanks!
> Jason
>
> --
> Jason M. Swails
> BioMaPS,
> Rutgers University
> Postdoctoral Researcher
>
>
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber

_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Tue Feb 25 2014 - 14:30:02 PST
Custom Search