[bug #13259] full analysis-high Te values -- March 25, 2009 - 21:01

URL:
  <http://gna.org/bugs/?13259>

                 Summary: full analysis-high Te values
                 Project: relax
            Submitted by: None
            Submitted on: Wednesday 03/25/2009 at 21:01 CET
                Category: relax's source code
                Severity: 3 - Normal
                Priority: 5 - Normal
                  Status: None
                 Privacy: Public
             Assigned to: None
         Originator Name: Pierre-Yves Savard
        Originator Email: pierre-yves.savard@xxxxxxxxxxxxx
             Open/Closed: Open
         Discussion Lock: Any
                 Release: Repository: 1.3 line
        Operating System: GNU/Linux

    _______________________________________________________

Details:

Hello,

I ran the full_analysis.py script and got strange results: for some residues,
Te values are two times greater than the correlation time of the protein (~ 10
ns).

In the code, the upper limit for Te values is set to 1.5 X the value of Tm,
so should be around 15 ns in my case).

I am using relax 1.3 (repository version) and I see these values of Te in the
aic directory of several rounds, running the prolate diffusion tensor. Hence,
is model-free model elimination set properly in the script 'full_analysis.py'
?

===============

I attached a few files to reproduce the bug. These are included in the
bzipped tar archive.

1.
R1, R2 and NOE data for residue A44.

2.
PDB including only residue A44.

3.
Sequence including only A44.

4.
Modified script so only models 'm1' and 'm8' are minimised, and the diffusion
tensor is not optimized (setting a fixed diffusion tensor with parameters
similar as when all 150 residues are used for optimisation; this is essential
to reproduce the bug).

5.
Directory 'prolate' including the 'init' round as well as 'round_1'.

The bug should be reproduced with the script, creating sub-directory
'round_2' including 'aic/results.bz2' which will, when the bug is present,
chose model 'm8' for A44 with a ts of ~20 ns.

The script takes less than 2 minutes to execute on a standard laptop.


Thank you very much !

PY



    _______________________________________________________

File Attachments:


-------------------------------------------------------
Date: Wednesday 03/25/2009 at 21:01 CET  Name:
bug_report_Te_full_analysis__P-Y_S.tar.bz2  Size: 18kB   By: None

<http://gna.org/bugs/download.php?file_id=5506>

    _______________________________________________________

Reply to this item at:

  <http://gna.org/bugs/?13259>

_______________________________________________
  Message sent via/by Gna!
  http://gna.org/