Hi,
By relaxation parameters, do you mean model-free parameters? You
should now be able to use relax to do the relaxation curve-fitting
without duplicate spectra and with Monte Carlo simulations for error
propagation. Maybe with the proper R1 and R2 errors, you'll get quite
different model-free results.
As for a test data set, do you mean for example published relaxation
data which can be put through relax and compared to the published
results? Or to see if you get reasonable results? By using the new
type of analysis from the d'Auvergne and Gooley, 2008 papers
(http://www.nmr-relax.com/refs.html) together with improved model-free
model selection, model-free model elimination, hugely improved
model-free optimisation, and a few other improvements (all described
in my PhD thesis at
http://eprints.infodiv.unimelb.edu.au/archive/00002799/ and at
http://www.amazon.com/Protein-Dynamics-Model-free-Analysis-Relaxation/dp/3639057627/ref=sr_1_1?ie=UTF8&s=books&qid=1228345232&sr=8-1),
I can guarantee you that you will struggle to reproduce literature
results ;)
One good test set - as it is similar to your system - is the
bacteriorhodopsin (1-36)BR fragment of Orekhov et al., 1999. You'll
need to talk to Vladislav Orekhov or Dmitry Korzhnev about getting
this data. It consists of 9 data sets (3 fields). I have it in text
form, but I'd prefer if you ask the original authors for permission (I
can then send you the nicely formatted text files). Hopefully once
relax has BMRB STAR format writing abilities, it will become mandatory
for relaxation data deposition prior publication as it is with
chemical shift assignments and PDB files so that it will be easier to
access relaxation data and dynamics results. Another system is the
OMP data from Gitti et al. (2005). The relaxation data is available
in their supplementary material. For both of these systems you could
compare to the published values. However I would recommend you
compare the values to those in my thesis. Orekhov's work is very
accurate and very similar, although not identical, to the values
obtained through relax with the 'full_analysis.py' script. For the
OMP data, this system displays a catastrophic failure of the current
methodologies and hence you should only compare to the results in my
thesis. Oh, the bacteriorhodopsin is fast whereas OMP is slow - just
because of the sizes of the systems.
I hope this helps,
Edward
On Wed, Dec 3, 2008 at 7:23 PM, Tyler Reddy <TREDDY@xxxxxx> wrote:
Hi Edward,
As it stands I haven't been able to get reasonable relaxation parameters
from
the 'relax' program. One suggestion from my project committee was to push
through a set of data that we already have good results for with this
program.
I wonder if there's a small data set that would work well for this? Ideally
one
that converges easily to save me time!
Tyler
_______________________________________________
relax (http://nmr-relax.com)
This is the relax-users mailing list
relax-users@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-users
Re: Test Data