Dear Edward, Thanks so much for your detailed mail regarding the steps involved in model-free analysis using single magnetic field data. To harness the potential of relax-GUI, I am seriously considering doing the experiments in another field.
Thanks for your support. with best regards, Dhanas Muthu On Jan 31, 2012, at 7:45 AM, Edward d'Auvergne wrote: > Dear Dhanas, > > Welcome to the relax mailing lists. Thank you, but note that this > program is the work of a number of relax developers. You are correct > that you cannot use the GUI for single field strength data. For such > data, you have to revert to the ancient methodology in the Mandel, > Akke, and Palmer 1995 paper. This will require scripting, but > fortunately most of the scripts are already there for you, just not > fully automated. > > To perform this old-style analysis, this will require an initial > diffusion tensor estimate which you can obtain from relax by setting > all spins to model m0 and optimising the diffusion tensors (once for > spherical diffusion, for the prolate spheroid, the oblate spheroid, > and the ellipsoid). This approximates the R1/R2 ratio methods, but > has the advantage of not having to convert between the >2000 Euler > angle conventions/symmetries, etc. which are different between > Modelfree4, Tensor2, Dasha, and relax. Note you cannot simply take a > tensor from one program and use it in another, as all programs use > different Euler or spherical angle notations. There is no sample > script for this, so you will need to write it yourself. > > The next part is iterative. You need to fix the diffusion tensor and > optimise the model-free models (i.e. the mf_multimodel.py script). > Then you need to select the best model (the > sample_scripts/model_free/modsel.py script). And finally you have to > optimise the diffusion tensor again (this will require your own script > as there is no equivalent sample script). Using this final diffusion > tensor as a starting point, you need to repeat all the steps of > model-free model optimisation, model selection, and tensor > optimisation until convergence (identical chi-squared values for the > final optimisation). > > What you'd like to do will require script editing and writing on you > behalf. You will need extensive use of the user function > documentation in the manual. I do not believe anyone has used relax > to perform such an analysis for a long, long time (I could be wrong > though). Most scripts for this would be for the old relax version > 1.0.x or earlier, so they cannot be used with modern versions of > relax. This Mandel et al methodology is heading towards being two > decades old - it is a very ancient technique with many known and > documented problems which you will need to be very aware of! I > suggest starting with my review: > > 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) > > Or my paper: > > 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). > > The review covers most of the problems, and the 2008 paper shows an > example where such methodology clearly fails. To perform a single > field strength analysis, you need to be acutely aware of these > problems. If you have any problems with all of the scripting, please > don't hesitate to ask on the mailing list. I hope this helps. > > Regards, > > Edward > > > P. S. Collecting data at 2 or more field strengths (essentially with > both per-experiment temperature calibration and temperature > compensation) will allow you to avoid scripting and use the automated > analysis via the GUI. > > > > > On 30 January 2012 19:53, Dhanasekaran Muthu <[email protected]> wrote: >> First, I wanted to thank for creating such a wonderful program for mode-free >> analysis. Using the recent version with GUI, I was able to get T1, T2 and >> NOE values without much effort. But I am not able to proceed further to do >> the mode-free analysis because the relax needs more than one filed frequency >> data. My problem is that I have data from only one frequency and I do not >> know the sequence of steps involved in running the relax. I learned from >> the forum that this has been asked several times and answered several times. >> I am also aware of problems with single frequency data. But I have to work >> with only one frequency data for the time being. Even after reading the >> forum and well-detailed manual, I am not able to understand what are the >> steps I should be doing to get model-free data. My understanding so far is >> i) first run mf-multimodel.py and then ii) modesel.py after removing the >> models M4 to M9. It would be nice if some one post the details of how to use >> relax for single frequency data, if they were successful already running the >> calculations. I would like to connect with those people who are in my >> situation and created successful code for running relax. >> >> Thanks, >> >> Dhanas Muthu >> _______________________________________________ >> relax (http://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 _______________________________________________ relax (http://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
