Dear Peixiang.
I have just discovered a bug in relax, where the 'relax_times' is not
properly loaded from the data-store and sent to the target function of
model NS R1rho 2-site.
This means that currently 'NS R1rho 2-site' can only be used for fixed
time at the moment.
This bug has no influence
Hi,
For a permanent reference, here are the timings of all of the
dispersion models in relax, with statistics from 10 repetitions, of
the calculation time for 100 function calls for 100 spins. I.e.
either 10,000 function calls for single spins or 100 function calls
for the cluster of 100 spins.
Hi Troels,
For reference for the relax users reading this, the abbreviations
Troels used for the relaxation dispersion models can be decoded using
the relax wiki page
http://wiki.nmr-relax.com/Category:Relaxation_dispersion.
> Just a little heads-up.
>
> Within a week or two, we should be able to
Dear Andrew.
Thank you for your valuable comments!
It becomes clear for me, that I should sit down with theory to grasp,
why I actually do:
1) matrix exponential
2) matrix power
in all the numerical models.
Then I would be able to select which best method to apply.
For:
ns_cpmg_2site_star.py =
Hi Troels,
If you're dealing almost entirely with 2x2 matricies, ie:
R=[(a_11,a_12),
(a_21,a_22)]
You can express the eigenvalues and eigenvectors exactly in terms of
these coefficients:
eg: if J is the matrix of eigen vectors and R_D is the diagonal matrix
of eigenvalues such that:
R
>I am looking for a solution I can apply to all numerical models :-)
The 'close formed' solutions I describe are eigenvalue decompositions.
So that's just a slightly smart way of doing that rather than asking a
minimiser to do it for you. No general closed form solutions exist for
the eigenval
http://www.cs.cornell.edu/cv/researchpdf/19ways+.pdf
On 26/06/2014 09:00, Troels Emtekær Linnet wrote:
Dear Peixiang, Dear Andrew.
Just a little heads-up.
Within a week or two, we should be able to release a new version of relax.
This update has focused on speed, recasting the data to highe
Hi Andrew.
I am looking for a solution I can apply to all numerical models :-)
The current implementation, is the eigenvalue decomposition.
http://svn.gna.org/svn/relax/branches/disp_spin_speed/lib/dispersion/matrix_exponential.py
It the bottom, I provide the profiling for NS R1rho 2site.
You wi
Thank you Andrew.
I already looked through it. :-)
They don't really come to a conclusion, but suggest method 3. (page 10)
Do you have any experience with it?
Best
Troels
2014-06-26 10:14 GMT+02:00 Andrew Baldwin :
> http://www.cs.cornell.edu/cv/researchpdf/19ways+.pdf
>
>
>
>
> On 26/06/201
Dear Peixiang, Dear Andrew.
Just a little heads-up.
Within a week or two, we should be able to release a new version of relax.
This update has focused on speed, recasting the data to higher
dimensional numpy arrays, and
moving the multiple dot operations out of the for loops.
So we have quite m
Hi Peixiang,
Actually, for comparison purposes for applying the 'NS R1rho 2-site'
model (http://wiki.nmr-relax.com/NS_R1rho_2-site) to variable-time
R1rho-type data, Art Palmer's MP05 model would be much better
(http://wiki.nmr-relax.com/MP05) than the DPL94 model
(http://wiki.nmr-relax.com/DPL94)
Hi Peixiang,
Please see below:
> Congratulations about the new version of 3.2.2, I tried, it works well :)
Cheers. If you notice any other problems or strange behaviour, please
don't hesitate to submit a bug report
(https://gna.org/bugs/?func=additem&group=relax). Then that problem
will likel
Hi, all,
Congratulations about the new version of 3.2.2, I tried, it works well :)
Still one question of using the different relaxation time periods.
My R1rho RD experiment has different relaxation time periods, I could
input all the peaks by the loop.
Then I fit with 'NS 2-site R_1$\scripts
Hi Peixiang.
I am glad to see your interest in relax.
I joined the project around a year ago, since I know that fitting
CPMG and R1rho data in our lab, can be quite a struggle for our students.
So relax has come as a relief, where the GUI can get most things done.
But it seems very much, that yo
Hi Peixiang,
For your bug report, I am currently downloading and installing a Fedora 18
VM image to test and see if I can replicate the problem. As for the
problem you see with the script, that is because the metadata is not all
set up for the R2eff data. Please have a look at the script at
./te
Hi Peixiang,
If you do only have R2eff data, i.e. there is no original peak lists,
I'll assume that you would fit into the category of 'power user'. In
this case in the GUI, instead of using the peak intensity wizard via
the spectra list 'Add' button, try the "User
functions->relax_disp->r2eff_re
Hi Peixiang,
Do you have original peak lists with the peak intensities? Or only
processed R2eff data?
For the variable time dispersion data, if you have peak lists, this is
very easy to use in the relax GUI. From the perspective of the user,
there is actually no difference between the variable
Thanks, Troels,
I would say I did not find a better way than getting 'dirty' hands to
handle my data.
My data is not fix-time R1rho. In gui, I did not figure out where to
input the variable-relaxation-time spectra data.
I only find that the scripts M61 could do it easily. So I chose the M61
Hi Peixiang.
It sounds like you would like to get your hands "dirty" with code.
1) Regarding Global fitting / Clustering.
First perform a normal analysis, which makes a result folder for each
model, and a final model.
Then perform global analysis. Cluster residues you find interesting.
Point to
Thanks, Troels,
That is useful. when I first looked at this folder, it was really
confusing for me.
I am creating my own scripts, it starts to run. If it works, I could
post it and maybe try with Kjaergaard dataset and continue your tutorial.
Best,
Peixiang
On 05/22/2014 05:44 PM, Troels Emt
Thanks, Edward and Troels,
'The trick with the auto-analysis is to use the result of the analytic
model as the starting point for optimisation of the numeric model, so
that the grid search is bypassed.'
That is very useful trick, I solved most of the errors with it.
I did not mean 8 inc is h
Hi Peixiang.
I started the R1rho tutorial at the wiki, which I think you refer to:
http://wiki.nmr-relax.com/Tutorial_for_Relaxation_dispersion_analysis_r1rho_fixed_time_recorded_on_varian_as_sequential_spectra
But I did not have the time to finish it.
But, if you are a little adventurous, you c
Hi Peixiang.
For model selection, try this in relax.
The help section is the best you can get !
Just start relax in terminal:
> relax
Then write:
relax> help(relax_disp)
to see all options.
For example:
help(relax_disp.select_model)
You can find more info at the wiki:
http://wiki.nmr-relax.com/
Hi Peixiang,
You should find the answers to most of your questions in the relax
manual. Please see below:
> Thanks for you reply. I used the script mode in linux.
>
> I tried with 'r1rho_on_res_m61.py', but I think in the script it has a
> mistake.
> in line27 ds.fixed = True, I think it should
Hi, Edward,
Thanks for you reply. I used the script mode in linux.
I tried with 'r1rho_on_res_m61.py', but I think in the script it has a
mistake.
in line27 ds.fixed = True, I think it should be False, because this
experiment has different time period, they do not need the ref. I
changed to F
Hi Peixiang,
Welcome to the relax mailing lists! The relaxation dispersion
analysis implemented in relax is quite flexible, and the data you have
is supported. This is well documented in the relax manual which you
should have with your copy of relax (the docs/relax.pdf file). Have a
look at sec
26 matches
Mail list logo
