Hi Atul,
As Troels asked, could you create a support request with the link
https://gna.org/support/?func=additem&group=relax and attach the files
there? Files should not be attached to public mailing list messages
as it puts a large amount of strain on the infrastructure, not only
with the mail server but also the 4 independent mailing list archives:
https://mail.gna.org/public/relax-users/2014-08/msg00025.html
http://www.mail-archive.com/relax-users@xxxxxxx/msg01672.html
http://thread.gmane.org/gmane.science.nmr.relax.user/1735/focus=1736
http://marc.info/?l=relax-users&r=1&w=2
As you can see from these archived messages, your attachment has been
stripped out of your email. The relax mail server is configured to
delete all attachments before sending out the message.
For your problem, I think I might now understand the issue. But it
would still be useful to have the files attached to the support
request for future reference. If you would like to keep the data
private, then maybe delete all data in the files except for one or two
residues (this also helps for converting the data into a relax system
test). And you could also slightly randomise the peak intensity
values if you really want to keep the real data private.
The issue is that the ordering of relaxation dispersion data in relax
is as follows:
1) Experiment type (certain dispersion models handle mixed experiment
types, for example the MMQ models,
2) The spins of the cluster,
3) Magnetic field strength,
4) Offsets (for both R1rho and CPMG, though off-resonance CPMG models
such as in CATIA are yet to be added to relax),
5) Dispersion points (nu_CPMG and nu_1).
So for 5), you only have one point per offset. But to specify the
reference, this needs to be at the lower dispersion point level. So
for each offset value, you will need to have your reference and the
point. You need to make relax load the reference peak intensities
multiple times, once for each offset value. So maybe you need:
---------------------------------------------------------------------------------------------------------------------------------------------------------------------
data = [
['900MHz_reference', 'ref_off_reso_R1rho_ubi_1.list', 2100,
None, 0.320, 900.0e6, 90000],
['900MHz_2100', '2100_off_reso_R1rho_ubi_2.list', 2100,
1500, 0.320, 900.0e6, 90000],
['900MHz_reference', 'ref_off_reso_R1rho_ubi_1.list', 2728,
None, 0.320, 900.0e6, 90000],
['900MHz_2728', '2728_off_reso_R1rho_ubi_3.list', 2728,
1500, 0.320, 900.0e6, 90000],
['900MHz_reference', 'ref_off_reso_R1rho_ubi_1.list', 3357,
None, 0.320, 900.0e6, 90000],
['900MHz_3357', '3357_reso_R1rho_ubi_4.list', 3357,
1500, 0.320, 900.0e6, 90000],
['900MHz_reference', 'ref_off_reso_R1rho_ubi_1.list', 3985,
None, 0.320, 900.0e6, 90000],
['900MHz_3985', '3985_off_reso_R1rho_ubi_5.list', 3985,
1500, 0.320, 900.0e6, 90000],
['900MHz_reference', 'ref_off_reso_R1rho_ubi_1.list', 4614,
None, 0.320, 900.0e6, 90000],
['900MHz_4614', '4614_off_reso_R1rho_ubi_6.list', 4614,
1500, 0.320, 900.0e6, 90000],
['900MHz_rep_4614','4614_rep_off_reso_R1rho_ubi_10.list', 4614,
1500, 0.320, 900.0e6, 90000],
['900MHz_reference', 'ref_off_reso_R1rho_ubi_1.list', 5242,
None, 0.320, 900.0e6, 90000],
['900MHz_5242', '5242_off_reso_R1rho_ubi_7.list', 5242,
1500, 0.320, 900.0e6, 90000],
['900MHz_reference', 'ref_off_reso_R1rho_ubi_1.list', 5871,
None, 0.320, 900.0e6, 90000],
['900MHz_5871', '5871_off_reso_R1rho_ubi_8.list', 5871,
1500, 0.320, 900.0e6, 90000],
['900MHz_reference', 'ref_off_reso_R1rho_ubi_1.list', 6500,
None, 0.320, 900.0e6, 90000],
['900MHz_6500', '6500_off_reso_R1rho_ubi_9.list', 6500,
1500, 0.320, 900.0e6, 90000]
]
---------------------------------------------------------------------------------------------------------------------------------------------------
Please carefully check the printouts to make sure the correct numbers
are being sent into the correct user functions. Specifying the
reference at the higher offset level is not yet supported in relax.
But with your data set attached to a support request, maybe in the
future it could be turned into a relax system test and such a feature
implemented. I hope this information helps.
Regards,
Edward
On 22 August 2014 23:01, Atul Srivastava <asrivast@xxxxxxx> wrote:
Dear Edward,
Thanks for providing help and insights. I used an improved script. The
scripts is able to calculate the R2eff and the omegaeffective properly, but
it is not able to fit any model. It also sees dispersion point as a constant
1500 Hz (see attached file in results/R2eff/disp_55_N.out).
I have attached all sample data, script and results for your review. Please
see the last few lines in the log file for the error message.
Thanks,
Atul
On Fri, Aug 22, 2014 at 3:02 AM, Edward d'Auvergne <edward@xxxxxxxxxxxxx>
wrote:
Hi Atul,
Again continuing from Troels' post at
http://thread.gmane.org/gmane.science.nmr.relax.user/1718/focus=1732.
Well, actually, as Troels fully covered all of you questions in more
detail than I could have provided, I don't have anything to add. If
you have other questions, please don't hesitate to ask.
Regards,
Edward
On 21 August 2014 12:00, Troels Emtekær Linnet <tlinnet@xxxxxxxxxxxxx>
wrote:
[snip]
relax_disp.relax_time(spectrum_id=id, time=relax_time)
-> This is used for the initial R2eff calculation, for exponential
curve fitting.
# Set the NMR field strength of the spectrum.
spectrometer.frequency(id=id, frq=H_frq)
-> This is used for conversion between nucleus, etc.
# Load the R1 data.
relax_data.read(ri_id='500MHz', ri_type='R1', frq=500e6,
file='R1_500MHz.out', dir=DATA_PATH, mol_name_col=1, res_num_col=2,
res_name_col=3, spin_num_col=4, spin_name_col=5, data_col=6,
error_col=7)
-> R1 needs to be loaded. This is because that R1 is part of the
equations. :-) The next release of relax, will implement feature where
R1 is fitted.
Note, that fitting introduces another variable in the equations.
# Read the chemical shift data.
chemical_shift.read(file='ref_500MHz.list', dir=DATA_PATH)
-> This is needed to get the ppm position of the nucleus.
-> See now:
http://www.nmr-relax.com/manual/Dispersion_model_summary.html
-> The "average resonance in the rotating frame" Omega(bar) = w_(bar) -
w_rf."
-> Here w_bar is the chemical shift, and w_rf is the offset.
-> Let is review figure Fig1_Palmer_Massi_2006.png
-> This is the S_z axis.
That should be it.
If you have any questions, please don't hesitate to write again.
If you need more help, consider writing a support request on the
homepage tracker.
- https://gna.org/support/?group=relax
Add following information:
# Please attach a system info file
relax -i -t relax_i.txt
# Please write up, which buttons you pushed, or attach your script.
# Consider adding your data in "sample" format. Meaning that you
delete all confidential information from the files, and only have 1-2
residues left for testing.
If you write such a support request, it it easier to share script files,
and
help other users.
The benefits from such a support request is:
- The information is available to all users, which can benefit
others in same situation.
- The information can be tracked back.
- The relax manual can be expanded, to help future users in same
situation.
Good luck!
Best
Troels Emtekær Linnet
PhD student
Copenhagen University
SBiNLab, 3-0-41
Ole Maaloes Vej 5
2200 Copenhagen N
Tlf: +45 353-22083
Lync Tlf: +45 353-30195
Re: field strength off-resonance R1rho constant relax time relaxation dispersion