Re: [relax-announce] relax version 4.1.3.

2019-07-12 Thread Stefano Luciano Ciurli
Hi Edward,
indeed, no models have been selected for any of the residues. That might be, 
logically, the problem. This is the message I read after each of the nine 
attempted models:

"RelaxWarning: No spins are selected therefore the optimisation or calculation 
cannot proceed."

Suggestions?
Stefano

> On 12 Jul 2019, at 15:57, Edward d'Auvergne  wrote:
> 
> On Fri, 12 Jul 2019 at 15:16, Stefano Luciano Ciurli
>  wrote:
>> 
>> Hi Edward,
>> I hope I am sending the message to the correct address now.
>> 
>> I think I have solved the issue that was causing the mentioned problem with 
>> the apparently incomplete set up by doing what you indicated.
>> 
>> Now the calculation has gone for about 5 min with the following result:
>> 
>> ==
>> = Completion of the d'Auvergne protocol model-free auto-analysis =
>> ==
>> 
>> Elapsed time:  5 minutes and 54.350 seconds
>> 
>> Exception raised in thread.
>> 
>> Traceback (most recent call last):
>>  File "gui/analyses/execute.pyc", line 87, in run
>>  File "gui/analyses/auto_model_free.pyc", line 834, in run_analysis
>>  File "auto_analyses/dauvergne_protocol.pyc", line 249, in __init__
>>  File "auto_analyses/dauvergne_protocol.pyc", line 610, in execute
>>  File "auto_analyses/dauvergne_protocol.pyc", line 855, in model_selection
>>  File "prompt/uf_objects.pyc", line 161, in __call__
>>  File "pipe_control/model_selection.pyc", line 260, in select
>>  File "pipe_control/pipes.pyc", line 68, in bundle
>>  File "lib/checks.pyc", line 81, in __call__
>> RelaxNoPipeError: RelaxError: The data pipe 'aic - mf (Sun Jul  7 18:21:56 
>> 2019)' has not been created yet.
>> 
>> What should I make of it?
> 
> Cheers!  The error message has been seen before:
> 
>
> http://www.nmr-relax.com/mail.gna.org/public/relax-users/2012-11/msg5.html
>
> http://www.nmr-relax.com/mail.gna.org/public/relax-users/2008-11/msg00034.html
> 
> However I've never had a bug report with enough details to reproduce
> the problem to be able to catch it in the test suite.  My guess is
> that if you carefully look at the log messages (hopefully you ran it
> with the log output to file), you'll see a series of RelaxWarning
> messages earlier which explain the problem in detail.  I would also
> guess that no models have been select for one or more spin systems.
> 
> Regards,
> 
> Edward



___
nmr-relax-announce mailing list
nmr-relax-announce@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nmr-relax-announce


Re: [relax-announce] relax version 4.1.3.

2019-07-12 Thread Edward d'Auvergne
On Fri, 12 Jul 2019 at 15:16, Stefano Luciano Ciurli
 wrote:
>
> Hi Edward,
> I hope I am sending the message to the correct address now.
>
> I think I have solved the issue that was causing the mentioned problem with 
> the apparently incomplete set up by doing what you indicated.
>
> Now the calculation has gone for about 5 min with the following result:
>
> ==
> = Completion of the d'Auvergne protocol model-free auto-analysis =
> ==
>
> Elapsed time:  5 minutes and 54.350 seconds
>
> Exception raised in thread.
>
> Traceback (most recent call last):
>   File "gui/analyses/execute.pyc", line 87, in run
>   File "gui/analyses/auto_model_free.pyc", line 834, in run_analysis
>   File "auto_analyses/dauvergne_protocol.pyc", line 249, in __init__
>   File "auto_analyses/dauvergne_protocol.pyc", line 610, in execute
>   File "auto_analyses/dauvergne_protocol.pyc", line 855, in model_selection
>   File "prompt/uf_objects.pyc", line 161, in __call__
>   File "pipe_control/model_selection.pyc", line 260, in select
>   File "pipe_control/pipes.pyc", line 68, in bundle
>   File "lib/checks.pyc", line 81, in __call__
> RelaxNoPipeError: RelaxError: The data pipe 'aic - mf (Sun Jul  7 18:21:56 
> 2019)' has not been created yet.
>
> What should I make of it?

Cheers!  The error message has been seen before:


http://www.nmr-relax.com/mail.gna.org/public/relax-users/2012-11/msg5.html

http://www.nmr-relax.com/mail.gna.org/public/relax-users/2008-11/msg00034.html

However I've never had a bug report with enough details to reproduce
the problem to be able to catch it in the test suite.  My guess is
that if you carefully look at the log messages (hopefully you ran it
with the log output to file), you'll see a series of RelaxWarning
messages earlier which explain the problem in detail.  I would also
guess that no models have been select for one or more spin systems.

Regards,

Edward


___
nmr-relax-announce mailing list
nmr-relax-announce@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nmr-relax-announce


Re: [relax-announce] relax version 4.1.3.

2019-07-12 Thread Stefano Luciano Ciurli
Hi Edward,
I hope I am sending the message to the correct address now.

I think I have solved the issue that was causing the mentioned problem with the 
apparently incomplete set up by doing what you indicated.

Now the calculation has gone for about 5 min with the following result:

==
= Completion of the d'Auvergne protocol model-free auto-analysis =
==

Elapsed time:  5 minutes and 54.350 seconds

Exception raised in thread.

Traceback (most recent call last):
  File "gui/analyses/execute.pyc", line 87, in run
  File "gui/analyses/auto_model_free.pyc", line 834, in run_analysis
  File "auto_analyses/dauvergne_protocol.pyc", line 249, in __init__
  File "auto_analyses/dauvergne_protocol.pyc", line 610, in execute
  File "auto_analyses/dauvergne_protocol.pyc", line 855, in model_selection
  File "prompt/uf_objects.pyc", line 161, in __call__
  File "pipe_control/model_selection.pyc", line 260, in select
  File "pipe_control/pipes.pyc", line 68, in bundle
  File "lib/checks.pyc", line 81, in __call__
RelaxNoPipeError: RelaxError: The data pipe 'aic - mf (Sun Jul  7 18:21:56 
2019)' has not been created yet.

What should I make of it?
Stefano


> On 12 Jul 2019, at 09:22, Edward d'Auvergne  wrote:
> 
> On Sun, 7 Jul 2019 at 18:37, Stefano Luciano Ciurli
>  wrote:
>> 
>> Hi Edward,
>> 
>> thank you for this update. The previous problems are solved now!
>> 
>> Proceeding further, and having loaded spins from a PDB structure which 
>> includes, of course, H and N nuclei, I receive a message when executing the 
>> program that says that the set up is incomplete, and that interatomic data 
>> for the dipole-dipole interaction is missing, followed by the full list of N 
>> and H atoms for each residue in the protein sequence.
>> 
>> Also, it suggests to try the “spin-isotope user function”.
>> 
>> What should I do?
> 
> Hi Stefano,
> 
> It took a while for your message to get through as you emailed the
> relax announcement mailing list.  I've now changed the address for
> this email thread to the relax users mailing list.  I have configured
> the relax announcement list to set the explicit "Reply-To:" header to
> the users mailing list.  Could you please deactivate the setting in
> your email software that is causing this to be ignored?  This
> "Reply-To:" header is also used for the commits mailing list to direct
> those responses to the development mailing list.  The announcement and
> commit mailing lists are read-only.
> 
> For this "The set up is incomplete.  Please check for the following
> missing information: Interatomic data (for the dipole-dipole
> interaction)" issue, have you clicked on all of the buttons under the
> relaxation data list GUI element?  You need to click on each one of
> these buttons to complete the set up.  Note that these are
> deliberately not mandatory, as people with corner case molecular
> systems will sometimes instead use the user function menus to set up
> their non-standard spin systems.
> 
> Regards,
> 
> Edward


___
nmr-relax-announce mailing list
nmr-relax-announce@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nmr-relax-announce


Re: [relax-announce] relax version 4.1.3.

2019-07-12 Thread Edward d'Auvergne
On Sun, 7 Jul 2019 at 18:37, Stefano Luciano Ciurli
 wrote:
>
> Hi Edward,
>
> thank you for this update. The previous problems are solved now!
>
> Proceeding further, and having loaded spins from a PDB structure which 
> includes, of course, H and N nuclei, I receive a message when executing the 
> program that says that the set up is incomplete, and that interatomic data 
> for the dipole-dipole interaction is missing, followed by the full list of N 
> and H atoms for each residue in the protein sequence.
>
> Also, it suggests to try the “spin-isotope user function”.
>
> What should I do?

Hi Stefano,

It took a while for your message to get through as you emailed the
relax announcement mailing list.  I've now changed the address for
this email thread to the relax users mailing list.  I have configured
the relax announcement list to set the explicit "Reply-To:" header to
the users mailing list.  Could you please deactivate the setting in
your email software that is causing this to be ignored?  This
"Reply-To:" header is also used for the commits mailing list to direct
those responses to the development mailing list.  The announcement and
commit mailing lists are read-only.

For this "The set up is incomplete.  Please check for the following
missing information: Interatomic data (for the dipole-dipole
interaction)" issue, have you clicked on all of the buttons under the
relaxation data list GUI element?  You need to click on each one of
these buttons to complete the set up.  Note that these are
deliberately not mandatory, as people with corner case molecular
systems will sometimes instead use the user function menus to set up
their non-standard spin systems.

Regards,

Edward


___
nmr-relax-announce mailing list
nmr-relax-announce@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nmr-relax-announce


[relax-announce] relax version 4.1.3.

2019-06-14 Thread Edward d'Auvergne
This is a minor bugfix release that re-enables the reading of Bruker
Dynamics Center NOE data files.

For the official, easy to navigate release notes, please see
http://wiki.nmr-relax.com/Relax_4.1.3 .

The new relax versions can be downloaded from
http://www.nmr-relax.com/download.html. If binary distributions are
not yet available for your platform and you manage to compile the
binary modules, please consider contributing these to the relax
project (described in section 3.6 of the relax manual,
http://www.nmr-relax.com/manual/relax_distribution_archives.html).

The full list of changes is:

Features:
N/A


Changes:
* FSF Copyright Validation configuration: Blacklisted the PDF user
manual.  This allows the checking of relax tags to pass.
* Release checklist document: Describe the relax fork of latex2html.
* API manual: No longer raise errors when parsing the pystarlib docstrings.
* Release checklist document: Minor improvements to match the
practical aspects of the release.
* User manual: Proper abbreviation of the "Quarterly Reviews of
Biophysics" journal name.
* Test suite: New system test to catch the failure of reading
newer Bruker DC NOE data files.  The system test is
Bruker.test_bug_15_NOE_read_fail and it catches bug #15
(https://sourceforge.net/p/nmr-relax/tickets/15/).  The test uses
truncated data from Stefano Ciurli as attached to the bug report.
* Bruker DC: Silence the warnings about spin names already
existing.  The user does not need to see such warnings.
* Travis CI config: Explicitly set 'trusty' as the distribution
name for the default images.  In the support request titled "Failure
of GUI testing via xvfb"
(https://support.travis-ci.com/hc/en-us/requests/7654), the Travis CI
support staff suggested that we explicitly set 'dist: trusty'.
* Bruker DC: A different way to silence the warnings about spin
names already existing.  The previous attempt at setting the force
flag to True was causing failures in a number of system tests.
Therefore a new flag 'warn_flag' has been added to
pipe_control.mol_res_spin.name_spin() to allow warnings to be
explicitly silenced.
* Travis CI config: Use Xenial for running all tests on Linux and
Python 2.7.  This is from the support request titled "Failure of GUI
testing via xvfb"
(https://support.travis-ci.com/hc/en-us/requests/7654).
* Travis CI config: Manual support for old SciPy versions on
Python 2.7.  SciPy 1.3.0 now requires Python >= 3.5.  Therefore the
OLD_MATPLOTLIB variable has been renamed to OLD_PY2_PACKAGES and, when
set, is now used to install old matplotlib and scipy versions when
using Python 2.7.
* Travis CI config: Deactivate the Mac OS X updates to avoid
timeouts.  The 'brew update' and 'brew upgrade python3' take up half
of the build time for the Mac OS X target.  This large amount of time
sometimes causes this build to hit the Travis CI time limits.


Bugfixes:
* Bruker DC: Support for handling newer versions of the NOE data
file.  This fixes bug #15
(https://sourceforge.net/p/nmr-relax/tickets/15/), the failure to read
newer versions of the Bruker DC NOE data files.  This was simply a
parsing issue as the NOE column is now "NOE [ ]" whereas previous DC
versions used the text "NOE" or "NOE [none]".


___
nmr-relax-announce mailing list
nmr-relax-announce@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nmr-relax-announce