bugrelax - Bugs: bug #20531, Fully automated model free...

 
 
Show feedback again

bug #20531: Fully automated model free calculations finish with the error "RelaxFault: RelaxError: Impossible to be here"

Submitted by:  None
Submitted on:  Tue 19 Feb 2013 04:13:40 PM UTC  
 
Category: relax's source codeSpecific analysis category: None
Priority: 5 - NormalSeverity: 3 - Normal
Status: FixedAssigned to: Edward d Auvergne <bugman>
Originator Name: James TolchardOriginator Email: -unavailable-
Open/Closed: ClosedRelease: 2.2.2
Operating System: All systems

Add a New Comment (Rich MarkupRich Markup):
   

You are not logged in

Please log in, so followups can be emailed to you.

 

(Jump to the original submission Jump to the original submission)

Wed 20 Feb 2013 02:48:17 PM UTC, SVN revision 18532:

Merged revisions 18531 via svnmerge from
svn+ssh://bugman@svn.gna.org/svn/relax/trunk

........
r18531 | bugman | 2013-02-20 15:47:45 +0100 (Wed, 20 Feb 2013) | 9 lines

Fix for bug #20531 (https://gna.org/bugs/?20531).

This is the bug of the RelaxFault during the final stages of the auto_analysis.dauvergne_protocol
model-free analysis when the Molmol and PyMOL macros are being created. The problem was caused by
model m9 (just the Rex parameter) when trying to determine what the order parameter for fast (< 200
ps) motions is. The code was assuming that the model-free model would always contain an order
parameter. It is very strange that this has only been rarely encountered!
........

(Browse SVN revision 18532)

Edward d Auvergne <bugman>
Project AdministratorIn charge of this item.
Wed 20 Feb 2013 02:47:45 PM UTC, SVN revision 18531:

Fix for bug #20531 (https://gna.org/bugs/?20531).

This is the bug of the RelaxFault during the final stages of the auto_analysis.dauvergne_protocol
model-free analysis when the Molmol and PyMOL macros are being created. The problem was caused by
model m9 (just the Rex parameter) when trying to determine what the order parameter for fast (< 200
ps) motions is. The code was assuming that the model-free model would always contain an order
parameter. It is very strange that this has only been rarely encountered!

(Browse SVN revision 18531)

Edward d Auvergne <bugman>
Project AdministratorIn charge of this item.
Wed 20 Feb 2013 02:37:59 PM UTC, SVN revision 18528:

Merged revisions 18527 via svnmerge from
svn+ssh://bugman@svn.gna.org/svn/relax/trunk

........
r18527 | bugman | 2013-02-20 15:37:20 +0100 (Wed, 20 Feb 2013) | 5 lines

The Mf.test_bug_20531_molmol_macro_write_relaxfault system test now catches bug #20531.

This now uses the results file attached to the bug report https://gna.org/bugs/?20531.
........

(Browse SVN revision 18528)

Edward d Auvergne <bugman>
Project AdministratorIn charge of this item.
Wed 20 Feb 2013 02:37:20 PM UTC, SVN revision 18527:

The Mf.test_bug_20531_molmol_macro_write_relaxfault system test now catches bug #20531.

This now uses the results file attached to the bug report https://gna.org/bugs/?20531.

(Browse SVN revision 18527)

Edward d Auvergne <bugman>
Project AdministratorIn charge of this item.
Wed 20 Feb 2013 02:29:46 PM UTC, comment #7:

Yes, that's fine. Many thanks!

James Tolchard <jtolchard>
Wed 20 Feb 2013 02:28:15 PM UTC, comment #6:

If you don't mind, I'll add this results file to the relax test suite. I'll remove all of the interatomic data containers as well (except for this residue :57).

Edward d Auvergne <bugman>
Project AdministratorIn charge of this item.
Wed 20 Feb 2013 02:13:39 PM UTC, comment #5:

I have gone through the troubleshooting as you described and it turns out 1 residue (the only m9) was responsible.

I attach the `relax --info` output and the minimum results file required to for your test to fail. I think I have successfully removed all identifying features, but if you think otherwise please let me know.

Cheers!

(file #17280, file #17281)

James Tolchard <jtolchard>
Wed 20 Feb 2013 09:35:25 AM UTC, comment #4:

I unfortunately cannot currently reproduce this bug. I created the system test:

$ relax -s Mf.test_bug_20531_molmol_macro_write_relaxfault

to try to trigger the bug, but that did not work!

A better way to do this would be if the final results file could be attached to this bug report (you should have the file final/results.bz2 at this stage). To keep your data confidential and to help in the debugging, would you be able to perform the following steps:

1) First, please attach the information from '$ relax --info', as this was unfortunately not attached last time. This can simply be pasted into the 'Post a Comment' section.

2) Download the attached 'bug_trigger.py' script into your model-free analysis directory.

3) Run it in the prompt mode with '$ relax bug_trigger.py'.

4) See that the RelaxFault occurs with this script.

5) Open the final/results.bz2 compressed XML file in a text editor and look for the residue sections.

6) Delete blocks of residues (from the <res ...> to </res> tags) until you are left with just one or two residues which trigger the RelaxFault. Make sure you preserve the <mol ...> and </mol> tags. Any mistakes you make will be very obvious when you run the 'bug_trigger.py' script.

7) Attach this minimum results file which triggers the bug. With only 1 or 2 residues, most of your data is kept confidential. You can even randomise the relaxation data and model-free parameter values of the remaining residues, and remove PDB references and structural data if you would like 100% confidentiality.

If you can do this, then with this absolute minimal relax state file I should be able to find a fix to the problem within 5 minutes. The trick is that I can add the state file, with your permission of course, to the relax test suite and create a quick system test which triggers the bug (based on the 'bug_trigger.py' script). Once I can reliably reproduce the bug in a system test, I can usually fix the bug in less time it takes to make the system test. And an additional advantage is that the bug can never return to relax. Cheers!

(file #17274)

Edward d Auvergne <bugman>
Project AdministratorIn charge of this item.
Wed 20 Feb 2013 08:54:00 AM UTC, SVN revision 18512:

Merged revisions 18511 via svnmerge from
svn+ssh://bugman@svn.gna.org/svn/relax/trunk

........
r18511 | bugman | 2013-02-20 09:49:10 +0100 (Wed, 20 Feb 2013) | 7 lines

Created the Mf.test_bug_20531_molmol_macro_write_relaxfault system test.

This is an attempt at catching bug #20531. It creates all of the m0-m9 and tm0-tm9 models, sets
some parameter values, and then attempts to create all of the Molmol macros, PyMOL macros, Grace
plots and parameter text files as present in the auto_analysis.dauvergne_protocol module.
........

(Browse SVN revision 18512)

Edward d Auvergne <bugman>
Project AdministratorIn charge of this item.
Wed 20 Feb 2013 08:49:10 AM UTC, SVN revision 18511:

Created the Mf.test_bug_20531_molmol_macro_write_relaxfault system test.

This is an attempt at catching bug #20531. It creates all of the m0-m9 and tm0-tm9 models, sets
some parameter values, and then attempts to create all of the Molmol macros, PyMOL macros, Grace
plots and parameter text files as present in the auto_analysis.dauvergne_protocol module.

(Browse SVN revision 18511)

Edward d Auvergne <bugman>
Project AdministratorIn charge of this item.
Wed 20 Feb 2013 08:25:57 AM UTC, comment #1:

This bug was first mentioned at http://marc.info/?l=relax-users&m=136126800616010&w=2.

Edward d Auvergne <bugman>
Project AdministratorIn charge of this item.
Tue 19 Feb 2013 04:13:40 PM UTC, original submission:

Using the method below for setting up a fully automated model free calculation in the relax 2.2.2 gui, the following error is given, although plausible model free parameters are determined:

Relax method.

1 - Start relax (./relax -g)
2 - Choose "New analysis", select Model-free analysis"
3 - Open the "Spin editor" and choose "From a new PDB structure"
4 - Select the PDB file (comprised of a single structural model) with "@H,N" spins, named, all other sections are left unset.
5 - Add R1, R2 and hNOE data sets at two magnetic fields (499685350 Hz and 800233758 Hz). All data sets calculated with relax. Set ID string, relax type, frequency and file name. All sets based on peak heights, with methanol calibration and single scan interleaving.
6 - Define the H isotope. Leave default values as 1H.
7 - Define the X isotope. Leave default values as 15N.
8 - Define CSA relaxation. Leave default values.
9 - Define Dipolar relaxation. Leave default values.
10 - "Execute relax"

Attached is the output from ./relax --info

Anonymous

 

(Note: upload size limit is set to 1024 kB, after insertion of the required escape characters.)

Attach File(s):
   
   
Comment:
   

Attached Files
file #17280:  relax_info.txt added by jtolchard (6kB - text/plain)
file #17281:  results_minimum_fail added by jtolchard (533kB - application/octet-stream)
file #17274:  bug_trigger.py added by bugman (3kB - text/x-python - This relax script is designed to reproduce the RelaxFault triggered within the auto_analysis.dauvergne_protocol module.)

 

Depends on the following items: None found

Items that depend on this one: None found

 

Carbon-Copy List
  • -unavailable- added by jtolchard (Updated the item)
  • -unavailable- added by bugman (Posted a comment)
  • -unavailable- added by None (Submitted the item)
  •  

    Do you think this task is very important?
    If so, you can click here to add your encouragement to it.
    This task has 0 encouragements so far.

    Only logged-in users can vote.

     

    Please enter the title of George Orwell's famous dystopian book (it's a date):

     

     

    Follow 10 latest changes.

    Date Changed By Updated Field Previous Value => Replaced By
    Wed 20 Feb 2013 02:48:45 PM UTCbugmanCategoryNone=>relax's source code
      StatusConfirmed=>Fixed
      Open/ClosedOpen=>Closed
      Operating SystemGNU/Linux=>All systems
    Wed 20 Feb 2013 02:28:15 PM UTCbugmanStatusIn Progress=>Confirmed
    Wed 20 Feb 2013 02:13:39 PM UTCjtolchardAttached File-=>Added relax_info.txt, #17280
      Attached File-=>Added results_minimum_fail, #17281
    Wed 20 Feb 2013 09:35:25 AM UTCbugmanAttached File-=>Added bug_trigger.py, #17274
      StatusNone=>In Progress
    Wed 20 Feb 2013 08:25:57 AM UTCbugmanAssigned toNone=>bugman
    Show feedback again

    Back to the top


    Powered by Savane 3.1-cleanup