I couldn't reproduce it either. Somehow it looks like a PDB file called '1shg-H.pdb' was being loaded, but from your messages I couldn't see where. I'll close the bug report, but it can be reopened later if the bug resurfaces.
Cheers, Edward On 30 July 2012 09:57, Martin Ballaschk <[email protected]> wrote: > Hi Edward, > > I couldn't reproduce the error with fresh data sets made on my linux machine. > So this seems to be non-issue in the end ... > > Cheers > Martin > > On 25.07.2012, at 14:42, Edward d'Auvergne <[email protected]> wrote: > >> Hi, >> >> This looks to be a genuine bug. If you could create a bug report, it >> would be much appreciated (see >> https://gna.org/bugs/?func=additem&group=relax). It would be useful >> to include the information from your mail. You could also attach the >> results file there, if you wish. Did you use the GUI for this, or a >> script for testing? It would be good to attach the output of running >> relax with teeing mode on (relax --tee log_file). >> >> An ideal case would be if you created a set of test files which only >> include information for one or two spins which still reproduce the >> problem. That way I'll be able to easily reproduce the bug and come >> up with a solution within a few minutes (hopefully). Without the test >> data, I would have to spend a few hours or days trying to come up with >> a data set which triggers the bug. I could then also include this in >> the relax test suite to ensure that the problem never returns. The >> test data can be randomised for confidentiality, if needed. >> >> Cheers, >> >> Edward >> >> >> >> On 25 July 2012 14:27, Martin Ballaschk <[email protected]> wrote: >>> >>> Hi Edward, >>> >>> thank you for your quick response. Thanks also for your help. >>> >>> There's still a problem when trying to setup the dipole pair, maybe I >>> messed something up: >>> >>>> dipole_pair.set_dist(spin_id1='@N', spin_id2='@H', ave_dist=1.02e-10) >>>> Traceback (most recent call last): >>>> File "gui/interpreter.pyc", line 299, in run >>>> File "generic_fns/dipole_pair.pyc", line 237, in set_dist >>>> File "generic_fns/interatomic.pyc", line 234, in interatomic_loop >>>> KeyError: '#1shg-H_mol1:7@N' >>> >>> I don't really know what that means. >>> >>> Here's what I set up: >>> >>> >>> 1. Read @N spins from a relaxation output file, via the spin editor, works >>> nicely: >>> >>> relax> sequence.read(file='rx_t1.out', dir=None, spin_id_col=None, >>> mol_name_col=None, res_num_col=2, res_name_col=3, spin_num_col=None, >>> spin_name_col=5, sep=None, spin_id=None) >>> Opening the file 'rx_t1.out' for reading. >>> # mol_name res_num res_name spin_num spin_name >>> None 6 LYS None N >>> . >>> . >>> . >>> >>> >>> 2. Attach protons: >>> >>> relax> sequence.attach_protons() >>> >>> >>> 3. Set isotope and element types: >>> >>> relax> spin.isotope(isotope='15N', spin_id='@N', force=True) >>> >>> relax> spin.isotope(isotope='1H', spin_id='@H', force=True) >>> >>> relax> spin.element(element='N', spin_id=':*@N', force=True) >>> >>> 4. Define NH pairs: >>> >>> relax> dipole_pair.define(spin_id1='@N', spin_id2='@H', direct_bond=True) >>> Magnetic dipole-dipole interactions are now defined for the following spins: >>> >>> # Spin_ID_1 Spin_ID_2 >>> ':6@N' ':6@H' >>> ':7@N' ':7@H' >>> . >>> . >>> . >>> >>> 5. set the distance between N and H in the NH: >>> >>> relax> dipole_pair.set_dist(spin_id1='@N', spin_id2='@H', ave_dist=1.02e-10) >>> Traceback (most recent call last): >>> File "gui/interpreter.pyc", line 299, in run >>> File "generic_fns/dipole_pair.pyc", line 237, in set_dist >>> File "generic_fns/interatomic.pyc", line 234, in interatomic_loop >>> KeyError: '#1shg-H_mol1:7@N' >>> >>> So I don't know where this comes from. The NH pairs are set up, and >>> dipole_pair.set_dist takes just the three parameters. Oddly, relax does not >>> complain about the first residue (6), but the second in the sequence (7). >>> From how it shows up in the spin editor, there is actually no difference >>> between the two. >>> >>> The relax state can be found here, if you want to have a look: >>> https://www.dropbox.com/s/heukkeycpwzsvmb/state2.bz2 >>> >>> I'd appreciate your support :) >>> >>> Regards >>> Martin >>> >>> >>> On 24.07.2012, at 16:20, Edward d'Auvergne wrote: >>> >>>> Hi Martin, >>>> >>>> In the new relax versions, you'll have to do things a little >>>> differently. In the case where no structure is available, you can >>>> just click on the 'Skip' buttons for the structural parts of the >>>> dipolar relaxation wizard. I was thinking about having a page at the >>>> start of the wizard to ask if structural data is available, and use >>>> this to skip the 2 structure related wizard pages (structure.read_pdb >>>> and structure.get_pos), but it's a lot of effort considering most >>>> users will have a 3D structure. Note that you'll also have to have >>>> both proton and nitrogen spins set up in the new relax versions to >>>> define the magnetic dipole-dipole interaction. This will allow >>>> multi-pole relaxation support to be added to relax in the future. >>>> >>>> I hope this info helps. You may encounter other problems in the GUI >>>> if you don't have a structure. The flexibility of the scripting mode >>>> is lost in the GUI, so it is hard to bring back all the features into >>>> the GUI. But relax will handle an analysis without a structure. >>>> >>>> Regards, >>>> >>>> Edward >>>> >>>> >>>> >>>> On 24 July 2012 16:05, Martin Ballaschk <[email protected]> wrote: >>>>> Hi Edward, >>>>> >>>>> I was wondering if it is possible to run relaxGUI's model-free routine if >>>>> there is no structure model available. Of course, only the local_tm or >>>>> spherical "global" models could be used for an analysis, then. >>>>> >>>>> In the current 2.1.0 release, the user is forced to load a structure file >>>>> ("dipolar relaxation" button). Shouldn't relax run without loading any >>>>> structural information if I select "sphere" or "local_tm" in the protocol >>>>> mode field? Looking at the sample scripts in the manual, this could >>>>> easily be done via a relax script (I haven't tried), but is it possible >>>>> also via the GUI? >>>>> >>>>> Cheers >>>>> Martin > > -- > Martin Ballaschk > AG Schmieder > Leibniz-Institut für Molekulare Pharmakologie > Robert-Rössle-Str. 10 > 13125 Berlin > [email protected] > Tel.: +49-30-94793-234/315 > Büro: A 1.26 > Labor: C 1.10 > _______________________________________________ relax (http://www.nmr-relax.com) This is the relax-users mailing list [email protected] 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-users
