Hello Matt--
>
> I am trying to finish up a new RNA structure using the 2016 python
> scripts. XPLOR-NIH version 2.53, installed on NMRBOX.
>
> I have been running the scripts in parallel, both the fold and
> refine scripts with no issue until today. I had just finished up a
> set of calculations, went through my nmr data to clear out noe
> violations, updated my NOE table and restarted my calcs.
>
> For some reason now, 3-4 of the structures in the run will no longer
> calculate and when I check log files, there is this error message:
>
> "xplorwrapproc_slave: Fatal error: slave timeout waiting for
> parent. 0 !-2775061"
This is an unusual failure- I wonder if there are any messages in the
syslog on the NMRbox server where you're running Xplor-NIH. Do you get
the same behavior with a different NMRbox server?
>
> This error only pops up for a couple structures )~4%) while the
> remaining run with no issue. This error is showing up on both the
> folding and refinement scripts and seems to scale with the number of
> structures I calculate. If I calculate 100, I have 4 failed
> structures, 200 total, 8 failures. Furthermore, the overall XPLOR
> run will no longer exit on its own and will not go through the
> statistics calculations.
>
You can modify your script a bit to do only the final analysis on the
structures you do calculate. For this, copy fold_TAR.py to analyze.py
and edit this file in the following fashion:
- remove the fixupCovalentGeom line
- edit the calcOneStructure function so that its body is empty (it
shouldn't do anything).
- to StructureLoop, remove the numStructures argument, and add this
argument:
pdbFilesIn="fold_Tar_*.pdb",
I'm happy to help further, and I would like to fix the root cause, but
it will probably take working with the NMRbox team.
best regards--
Charles
########################################################################
To unsubscribe from the XPLOR-NIH list, click the following link:
http://list.nih.gov/cgi-bin/wa.exe?SUBED1=XPLOR-NIH&A=1
