Re: [AMBER] MMPBSA.py decomp, -deo, PB only problem

From: Jason Swails <jason.swails.gmail.com>
Date: Fri, 26 Jul 2013 15:34:58 -0400

On Fri, Jul 26, 2013 at 11:29 AM, Ryan Pavlovicz <pavlovicz.7.osu.edu>wrote:

> Hi,
>
> I am experiencing a problem using the MMPBSA.py.MPI script when it comes to
> calculating the DELTA energies for every frame of a decomposition (-deo)
> when only PB calculations are called for.
>
> Here is my input file:
>
> &general
> endframe=4000, verbose=2, keep_files=2,
> /
> &pb
> istrng=0.150, inp=1, cavity_offset=0.92, cavity_surften=0.00542,
> /
> &decomp
> idecomp=1, print_res="1-255"
> dec_verbose=1,
> /
>
> And the output:
>
> Loading and checking parameter files for compatibility...
> sander found! Using /usr/local/amber/amber12/bin/sander
> cpptraj found! Using /usr/local/amber/amber12/bin/cpptraj
> Preparing trajectories for simulation...
> 4000 frames were processed by cpptraj for use in calculation.
>
> Running calculations on normal system...
>
> Beginning PB calculations with /usr/local/amber/amber12/bin/sander
> calculating complex contribution...
> calculating receptor contribution...
> calculating ligand contribution...
> UnboundLocalError: local variable 'gb_com' referenced before assignment
> Error occured on rank 0.
> Exiting. All files have been retained.
> [cli_0]: aborting job:
> application called MPI_Abort(MPI_COMM_WORLD, 1) - process 0
>
> This only occurs when I don't include the GB calculations. The -deo output
> will print the PB complex, receptor, and ligand info, but fails when it
> gets to the 'DELTAS:' section.
>
> I am experiencing this problem with the AmberTools version 13.13.
>
> This seems like a possible bug, so i thought i'd bring it to someone's
> attention.
>

It is. Can you change debug_printlevel to "2" in the _MMPBSA_info file and
then run:

MMPBSA.py -rewrite-output

? That should print out a verbose traceback that will let me find where
the problem is.

Thanks!
Jason

-- 
Jason M. Swails
BioMaPS,
Rutgers University
Postdoctoral Researcher
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Fri Jul 26 2013 - 13:00:02 PDT
Custom Search