search

google-code-export / nmrrestrntsgrid

Automatically exported from code.google.com/p/nmrrestrntsgrid
0 stars 0 forks source link

Processing failed at linking stage #102

Closed GoogleCodeExporter closed 9 years ago

GoogleCodeExporter commented 9 years ago 
This entry failed in postprocessing. No DOCR and FRED files generated. 
Here is the stacktracing in the log file. Jurgen believes this issue should
go to Wim. 

Model read finished. Duration 2.09325218201
Model validity check skipped
Read msd.adatah.localConstants.py version 0.0.3
Removing existing project...
Doing 1snl...
Traceback (most recent call last):
  File
"/big/docr/workspace/recoordD/python/recoord2/msd/linkNmrStarData.py", line
533, in ?
    LinkNmrStarData(sys.argv)
  File "/big/docr/workspace/recoordD/python/msd/adatah/Generic.py", line
87, in __init__
    self.catchError(raiseError,timeFlag)
  File "/big/docr/workspace/recoordD/python/msd/adatah/Generic.py", line
80, in __init__
    self.runSpecific()
  File
"/big/docr/workspace/recoordD/python/recoord2/msd/linkNmrStarData.py", line
158, in runSpecific
    self.writeNmrStarProject(outNmrStarFile,self.entry)
  File
"/big/docr/workspace/recoordD/python/recoord2/msd/linkNmrStarData.py", line
301, in writeNmrStarProject
    self.setNmrStarConversionErrorInfo()
  File
"/big/docr/workspace/recoordD/python/recoord2/msd/linkNmrStarData.py", line
404, in setNmrStarConversionErrorInfo
    originalConstraintItemId =
constraintItem.findFirstApplicationData(application = formatName, keyword =
'origId').value
AttributeError: 'NoneType' object has no attribute 'value'
~

Original issue reported on code.google.com by junf...@gmail.com on 14 Aug 2008 at 3:30

GoogleCodeExporter commented 9 years ago 
This is a complicated one - there were two problems with this entry.

1. The coordinate section starts at residue Leu-5, the sequence at Gly-1, but 
the
Leu-5 is numbered as 1 in author_seq_code (so in the PDB file). This causes two
problems: the coordinates are not being linked correctly, and the automatic 
mapping
of the constraints does not work because the FC can only handle custom 
numberings by
format name (i.e. NmrStar), and this file in effect has two numbering systems. 
This
is because of the switch to the 'joint' NMR-STAR format - before the 
constraints came
from NMR-STAR, the coordinates from PDB, and the differences in numbering were
handled automatically. Anyway, these issues are now resolved because I reset the
sequence codes for the residues that are not in the coordinate section, so that 
the
sequence numbering (in this case) starts at -3.

  1.1 Coordinate linking is now correct.
  1.2 Linking of sequence to resonances now works automatically (also this resolves
issue 105).

Note that this fix should improve a lot of entries.

2. The error that was reported is because some custom information was not set 
for RDC
restraints. This is now fixed.

These fixes should not interfere with entries that are handled correctly by the
current code. I've tested the code changes with several entries, but we might 
want to
do it on the set of 12 to be absolutely sure it all works.

Original comment by wfvran...@gmail.com on 19 Aug 2008 at 11:38

GoogleCodeExporter commented 9 years ago 
Thanks!
I'll do some more testing that I need to do for issue 107 anyway.

Original comment by jurge...@gmail.com on 19 Aug 2008 at 12:09

GoogleCodeExporter commented 9 years ago 
Can you tell me what to update? Use this template if you like:
1 would be required, zero is not.

1 ccpn source                  
0 ccpn api regeneration 
0 ccpn chemp comps 
0 recoord source 0
0 recoord presetDict.py only

Every relevant issue we assign to you we'll mention this template ok?

Original comment by jurge...@gmail.com on 19 Aug 2008 at 12:17

GoogleCodeExporter commented 9 years ago 
OK. This would be:

11010

API regen necessary because we just did an upgrade in stable last week.

Original comment by wfvran...@gmail.com on 19 Aug 2008 at 12:23

GoogleCodeExporter commented 9 years ago 
Thanks, perhaps easiest if you copy the list like below, but we can read it 
otherwise too.

1 ccpn source                  
1 ccpn api regeneration 
0 ccpn chemp comps 
1 recoord source 0
0 recoord presetDict.py only

I just merged the changes we had from last week into the recoord source for 
presetDict.py.
I hope this doesn't cause a problem on your side. I will always check to see if 
it's at least valid syntax before 
doing a weekly commit.

Original comment by jurge...@gmail.com on 19 Aug 2008 at 12:38

GoogleCodeExporter commented 9 years ago 
Verified for 1snl entry. Thanks Wim! 
I'll now turn on the entries for issue 107.

Original comment by jurge...@gmail.com on 19 Aug 2008 at 2:22