Re: first time doing model free
Molecular dynamics by NMR data analysis
Brought to you by:
edauvergne,
troelslinnet
Menu
▾
▴
Re: first time doing model free
|
From: Edward d'A. <edw...@do...> - 2012-05-11 09:34:58
|
Hi Fernando, Welcome to the relax mailing lists! Before I get into the error you have encountered, I have noticed that the analysis you have performed is far from complete. I will first detail the long list of things you will need to do to reimplement a full analysis using single field strength data, and all of the reading and learning you will have to do to be capable and competent enough to know when things go wrong in the analysis, and then you might see that measuring data at a second field strength will actually be simpler for you ;) In any case, to have a good understanding of the concepts, implementation and pitfalls of a model-free analysis, you will be required to do a lot of reading (see http://www.nmr-relax.com/refs.html for a few references, with links). You will know you are doing the right thing when your full set of calculations takes 1-2 weeks to complete! Firstly, I would recommend you read my previous responses to questions relating to single field strength data, and look the other messages in those threads. You will find these discussions quite informative and highly detailed: - Shantanu Bhattacharyya: http://www.mail-archive.com/rel...@do... - Dhanasekaran Muthu: http://www.mail-archive.com/rel...@do... - Aldino Viegas: http://www.mail-archive.com/rel...@do... - Pierre-Yves Savard: http://www.mail-archive.com/rel...@do... - Keith Constantine: http://www.mail-archive.com/rel...@do... - Clare-Louise Evans: http://www.mail-archive.com/rel...@do... - Hongyan Li: http://www.mail-archive.com/rel...@do... - Vitaly Vostrikov: http://www.mail-archive.com/rel...@do... I think that covers it. There might be a little bit of irrelevant information in there for you, but most of it will be exactly what you need. For information about the implementation that you will need to self-assemble for single field strength data, see: - Maddy Strickland: http://www.mail-archive.com/rel...@do... As for using single field strength data, you will need to reimplement a full analysis protocol. With the model-free.py sample script, you have only optimised the model-free parameters for a single model for all spin systems in your system. However this is far from complete. You will need to optimise for all model-free models (m0 to m5, http://www.nmr-relax.com/manual/model_free_models.html). This is done by the mf_multimodel.py sample script. After this you will need to perform model elimination, then model selection using AIC, and then optimise the global diffusion tensor. This this needs to be done iteratively, and this can take up to 15 iterations, see: d'Auvergne, E. J. and Gooley, P. R. (2008). Optimisation of NMR dynamic models I. Minimisation algorithms and their performance within the model-free and Brownian rotational diffusion spaces. J. Biomol. NMR, 40(2), 107-119. (http://dx.doi.org/10.1007/s10858-007-9214-2). You will then have to do this separately for all diffusion tensors (spherical, prolate spheroid, oblate spheroid, and ellipsoid). Note that you will have the problem of the initial diffusion tensor estimate, and you will need to know how and when this fails: d'Auvergne E. J., Gooley P. R. (2007). Set theory formulation of the model-free problem and the diffusion seeded model-free paradigm. Mol. Biosyst., 3(7), 483-494. (http://dx.doi.org/10.1039/b702202f) d'Auvergne, E. J. and Gooley, P. R. (2008). Optimisation of NMR dynamic models II. A new methodology for the dual optimisation of the model-free parameters and the Brownian rotational diffusion tensor. J. Biomol. NMR, 40(2), 121-133. (http://dx.doi.org/10.1007/s10858-007-9213-3) Another good reference which contains everything you'll ever need is my PhD thesis: d'Auvergne, E. J. (2006). Protein dynamics: a study of the model-free analysis of NMR relaxation data. Ph.D. thesis, Biochemistry and Molecular Biology, University of Melbourne. (abstract - http://eprints.infodiv.unimelb.edu.au/archive/00002799/, PDF - http://eprints.infodiv.unimelb.edu.au/archive/00002799/01/thesis.pdf, and my book form - http://www.amazon.com/Protein-Dynamics-Model-free-Analysis-Relaxation/dp/3639057627) For implementing a single field strength analysis, i.e. reverting back to how analyses were performed and published almost 2 decades ago, I would recommend a recent copy of the relax user manual (http://www.nmr-relax.com/manual/index.html) where I have completed the model-free analysis chapter (http://www.nmr-relax.com/manual/Model_free_analysis.html). The section titled 'The diffusion seeded paradigm' has a figure which details the iterative implementation that you will have to write into a script. There are full details about this in my PhD thesis if this is not enough. I would also recommend you be very careful with temperature control and you look at Sebastien Morin's relaxation data consistency testing built into relax. If you do not have temperature control (either via single scan interleaving or off-field temperature compensation pulses at the start, or both together) as well as per-experiment temperature calibration (running a shorted experiment on a MeOH or equivalent sample and seeing what the chemical shift and hence temperature is during an experiment), then with single field strength data you will fall into all of the traps of artificial nanosecond and millisecond motions. So maybe after reading about all of this, collecting data at a second field strength would really be easier ;) Regards, Edward P. S. The error you encountered is because the calculation you ran was not a full analysis - rather only a very tiny shard of a massive analysis. The extraction script from Michael Bieri will only operate on the final results of a full analysis. But note that it too requires an analysis using data at minimally two field strengths. On 10 May 2012 01:46, Fernando Amador <Fer...@do...> wrote: > Hi, > > I wanted to use relax to do some model-free analysis, unfortunately I do not > have data from multiple fields. Can I still use relax for model free > analysis? In most of the scripts I read it says data from multiple fields > is required and the gui version won't even let me run the script without > data from multiple fields. I figured I would still give it a shot so I > generated R1, R2 and steady state NOE output files using the relax gui and > then modified the model-free.py script by commenting out the lines for the > second field strength. The script runs without problems from the prompt and > it generates a results.bz2 file. I tried to run the > final_data_extraction.py script but I get this error > > relax> pipe.create(pipe_name='Data_extraction', pipe_type='mf') > relax> results.read(file='results', dir=None) > Opening the file 'results.bz2' for reading. > Traceback (most recent call last): > File "/sw/bin/relax", line 381, in <module> > Relax() > File "/sw/bin/relax", line 117, in __init__ > self.interpreter.run(self.script_file) > File "/sw/lib/relax-py27/prompt/interpreter.py", line 294, in run > return run_script(intro=self.__intro_string, local=locals(), > script_file=script_file, quit=self.__quit_flag, > show_script=self.__show_script, raise_relax_error=self.__raise_relax_error) > File "/sw/lib/relax-py27/prompt/interpreter.py", line 597, in run_script > return console.interact(intro, local, script_file, quit, > show_script=show_script, raise_relax_error=raise_relax_error) > File "/sw/lib/relax-py27/prompt/interpreter.py", line 483, in > interact_script > exec_script(script_file, local) > File "/sw/lib/relax-py27/prompt/interpreter.py", line 376, in exec_script > runpy.run_module(module, globals) > File "/sw/lib/python2.7/runpy.py", line 180, in run_module > fname, loader, pkg_name) > File "/sw/lib/python2.7/runpy.py", line 72, in _run_code > exec code in run_globals > File "/Users/Joe/NMR/relax/m4/final_data_extraction.py", line 36, in > <module> > spin_no = spin_id[spin_id.index(':')+1:spin_id.index('&')] > ValueError: substring not found > > I don't know if my initial model-free.py script worked. If it did, I would > like to look at the data. If relax can't analyze single field data, are > there any other programs that you would recommend. Any help would be > appreciated. > > > Thanks, > > Fernando > > > _______________________________________________ > relax (http://www.nmr-relax.com) > > This is the relax-users mailing list > rel...@do... > > 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 |
