Re: [bug #20213] duplicate_data / deepcopy: IndexError: list assignment index out of range -- October 04, 2012 - 18:05

Hi,

For attachments, could I ask you to avoid sending them to public
mailing lists?  This causes significant strain on the infrastructure.
It is better for reference to attach it to a relevant bug/support/task
report.

I can now reproduce the problem and see what the issue is.  If it is
ok, I will add your data in the files 'sh3-47.2.zip' and 'run.py'
(attached to https://gna.org/bugs/?20213) to the relax test suite so
that this bug will never be able to return to relax.  I can speed this
up for the test suite by only using model 'tm0', using 2 grid search
increments, and only having residue 47 in the PDB file.  The issue is
because of the following user function (with output):

relax> structure.load_spins(spin_id='@N*', ave_pos=True)
Adding the following spins to the relax data store.

# mol_name     res_num    res_name    spin_num    spin_name
1U06-H_mol1    47         ASN         350         N
1U06-H_mol1    47         ASN         360         ND2
1U06-H_mol1    47         ASN         361         ND2

This should in reality not be a problem - relax should handle this.
But a workaround which does not trigger the bug is to load only the
backbone nitrogens with:

structure.load_spins(spin_id='@N', ave_pos=True)

You will see this in the sample script for relax 2.1.1.  I will work
on this bug, but the temporary solution is to not load the Asn and Gln
sidechain nitrogens.  Loading the Trp sidechain is however ok.

Cheers,

Edward




On 4 October 2012 15:27, Martin Ballaschk <ballaschk@xxxxxxxxxxxxx> wrote:
> Hi Edward,
>
> this appears indeed to be completely unrelated to what I posted earlier.
>
> I attached the complete script "run.py" and relaxation files containing 
> only spin 47 to the bug report. Spin 47 alone is enough to bring relax 
> down. I can reproduce the error with a linux machine running Suse with / 
> without multi processor mode, and with 2.1.1 on another machine – although 
> I will re-test that now.
>
> Some guessing (no idea, if it's relevant): It is possible, that residue 47 
> shows behaviour that differs from the rest of the protein. Judging from the 
> B-factors of the included pdb file, it shows increased mobility. In the 
> relaxation experiments however I could see residue 48 showed high R2 rates, 
> and also having some Rex in the mf analysis (although the absolute value 
> seems very low) - see also the attached pdf file.
>
> Cheers
> Martin
>
> On 04.10.2012, at 14:01, "Edward d'Auvergne" <edward@xxxxxxxxxxxxx> wrote:
>
> > Hi Martin,
> >
> > I'm not currently able to reproduce the bug.  Would you have the full
> > script to reproduce this?  This appears to not be related to the
> > missing atoms in the PDB file problem you described at:
> >
> > http://thread.gmane.org/gmane.science.nmr.relax.user/1314
> > http://article.gmane.org/gmane.science.nmr.relax.user/1314
> >
> > There is no relaxation data for residues 6 and 62 in the files
> > attached to the bug report (https://gna.org/bugs/?20213).  Have you
> > tried deleting other data points until you have only 2-3 residues
> > which still trigger the bug?  To debug this, I would like to add the
> > absolute minimal data set and script into the relax test suite.  This
> > should then allow me to find the bug within minutes.  Without it, I am
> > currently a bit stuck.
> >
> > Cheers,
> >
> > Edward
> >
> >
> >
> >
> > On 2 October 2012 17:48, Martin Ballaschk
> > <NO-REPLY.INVALID-ADDRESS@xxxxxxx> wrote:
> > > URL:
> > >  <http://gna.org/bugs/?20213>
> > >
> > >                 Summary: duplicate_data / deepcopy: IndexError: list
> > > assignment index out of range
> > >                 Project: relax
> > >            Submitted by: mab
> > >            Submitted on: Tue 02 Oct 2012 03:48:51 PM GMT
> > >                Category: GUI
> > >                Severity: 3 - Normal
> > >                Priority: 5 - Normal
> > >                  Status: None
> > >                 Privacy: Public
> > >             Assigned to: None
> > >         Originator Name:
> > >        Originator Email:
> > >             Open/Closed: Open
> > >         Discussion Lock: Any
> > >                 Release: 2.1.0
> > >        Operating System: GNU/Linux
> > >
> > >    _______________________________________________________
> > >
> > > Details:
> > >
> > > Hi Edward,
> > >
> > > during or after fitting tm, relax gives me the following error:
> > >
> > > File "/opt/relax-2.1.0/specific_fns/model_free/main.py", line 1537, in
> > > duplicate_data
> > >    dp_to.mol[mol_index].res[res_index].spin[spin_index] = deepcopy(spin2)
> > > IndexError: list assignment index out of range
> > >
> > > The full traceback:
> > >
> > > Model-free model of spin '#1U06-H_mol1:47@N'.
> > > Data pipe            Num_params_(k)       Num_data_sets_(n)    Chi2
> > >     Criterion
> > > tm0 - mf (Tue Oct  2 17:14:03 2012) 1                    6
> > > 1051.35716           1053.35716
> > > tm1 - mf (Tue Oct  2 17:14:03 2012) 2                    6
> > > 28.84022             32.84022
> > > tm2 - mf (Tue Oct  2 17:14:03 2012) 3                    6
> > > 25.50256             31.50256
> > > tm3 - mf (Tue Oct  2 17:14:03 2012) 3                    6
> > > 3.76695              9.76695
> > > tm4 - mf (Tue Oct  2 17:14:03 2012) 4                    6
> > > 3.76695              11.76695
> > > tm5 - mf (Tue Oct  2 17:14:03 2012) 4                    6
> > > 25.50256             33.50256
> > > tm6 - mf (Tue Oct  2 17:14:03 2012) 5                    6
> > > 25.50256             35.50256
> > > tm7 - mf (Tue Oct  2 17:14:03 2012) 5                    6
> > > 3.76695              13.76695
> > > tm8 - mf (Tue Oct  2 17:14:03 2012) 6                    6
> > > 3.76695              15.76695
> > > tm9 - mf (Tue Oct  2 17:14:03 2012) 2                    6
> > > 683.15687            687.15687
> > > The model from the data pipe 'tm3 - mf (Tue Oct  2 17:14:03 2012)' has 
> > > been
> > > selected.
> > > Exception raised in thread.
> > >
> > > Traceback (most recent call last):
> > >  File "/opt/relax-2.1.0/gui/analyses/execute.py", line 87, in run
> > >    self.run_analysis()
> > >  File "/opt/relax-2.1.0/gui/analyses/auto_model_free.py", line 803, in
> > > run_analysis
> > >    dauvergne_protocol.dAuvergne_protocol(pipe_name=self.data.pipe_name,
> > > pipe_bundle=self.data.pipe_bundle, results_dir=self.data.save_dir,
> > > diff_model=self.data.global_models, mf_models=self.data.mf_models,
> > > local_tm_models=self.data.local_tm_models, grid_inc=self.data.inc,
> > > diff_tensor_grid_inc=self.data.diff_tensor_grid_inc,
> > > mc_sim_num=self.data.mc_sim_num, max_iter=self.data.max_iter,
> > > conv_loop=self.data.conv_loop)
> > >  File "/opt/relax-2.1.0/auto_analyses/dauvergne_protocol.py", line 235, in
> > > __init__
> > >    self.execute()
> > >  File "/opt/relax-2.1.0/auto_analyses/dauvergne_protocol.py", line 562, in
> > > execute
> > >    self.model_selection(modsel_pipe=self.name_pipe('aic'), 
> > > dir=self.base_dir
> > > + 'aic')
> > >  File "/opt/relax-2.1.0/auto_analyses/dauvergne_protocol.py", line 769, in
> > > model_selection
> > >    self.interpreter.model_selection(method='AIC', modsel_pipe=modsel_pipe,
> > > bundle=self.pipe_bundle, pipes=self.pipes)
> > >  File "/opt/relax-2.1.0/prompt/uf_objects.py", line 219, in __call__
> > >    self._backend(*new_args, **uf_kargs)
> > >  File "/opt/relax-2.1.0/generic_fns/model_selection.py", line 298, in 
> > > select
> > >    duplicate_data[best_model](best_model, modsel_pipe, model_info,
> > > global_stats=global_flag, verbose=False)
> > >  File "/opt/relax-2.1.0/specific_fns/model_free/main.py", line 1537, in
> > > duplicate_data
> > >    dp_to.mol[mol_index].res[res_index].spin[spin_index] = deepcopy(spin2)
> > > IndexError: list assignment index out of range
> > >
> > > The full error log is in log5.txt.
> > >
> > > I used a reduced data set of only 10 residues, still relax chokes on 
> > > residue
> > > 47.
> > >
> > > I have no idea where this comes from. With a different crystal structure 
> > > / pdb
> > > file (1shg) and relax version everything worked well and I got quite
> > > meaningful results. I wnated to re-run relax with another pdb structure, 
> > > just
> > > for the heck of it.
> > >
> > > Please have a look into the "prep.py" file, that's the script I used to
> > > prepare the setup.
> > >
> > > I'd be very grateful if you could tell me what went wrong.
> > >
> > > The recorded data are not confidential and I could make them accessible 
> > > at any
> > > time.
> > >
> > >
> > >
> > >
> > >    _______________________________________________________
> > >
> > > File Attachments:
> > >
> > >
> > > -------------------------------------------------------
> > > Date: Tue 02 Oct 2012 03:48:51 PM GMT  Name: log5.txt  Size: 613kB   By: 
> > > mab
> > >
> > > <http://gna.org/bugs/download.php?file_id=16619>
> > > -------------------------------------------------------
> > > Date: Tue 02 Oct 2012 03:48:51 PM GMT  Name: prep.py  Size: 3kB   By: mab
> > >
> > > <http://gna.org/bugs/download.php?file_id=16620>
> > >
> > >    _______________________________________________________
> > >
> > > Reply to this item at:
> > >
> > >  <http://gna.org/bugs/?20213>
> > >
> > > _______________________________________________
> > >  Message sent via/by Gna!
> > >  http://gna.org/
> > >
> > >
> > > _______________________________________________
> > > relax (http://www.nmr-relax.com)
> > >
> > > This is the relax-devel mailing list
> > > relax-devel@xxxxxxx
> > >
> > > To unsubscribe from this list, get a password
> > > reminder, or change your subscription options,
> > > visit the list information page at
> > > https://mail.gna.org/listinfo/relax-devel