Subject: Re: Pseudo 3D T2_relaxation
From: nickfitzkee
Date: Mar 13, 2011
Previous: 779
Hi Mohammed,
I could envision several solutions to your problem. The easiest is simply this: transform your spectra so that each plane corresponds to a single time point. Then, use pipe2ucsf on each of the ft2 files, which should allow you load the spectra individually and fit them in sparky. Depending on how youve set up your experiment, you may have to use the sethdr command to turn the pseudo-3D spectrum back into a collection of 2Ds.
A more sophisticated solution is to use NMRPipes modelXY command to fit your peaks. I prefer this method, because I like the noise estimation provided by modelXY, but the answers are generally comparable to Sparkys fit. In brief, you would use autoFit.tcl to create a table containing each peak intensity series:
autoFit.tcl -series -dX 0.0 -dY 0.0 -inTab (peak table)
-specName ft/test%03d.ft2 -nofit
Where (peak table) is the assigned peak table of your first plane. -series tells it to fit a pseudo-3D, -dN 0.0 tells it not to shift peak positions (the exact same matrix point is used in each spectra), and -nofit tells it not to bother with fitting line widths or intensities. The output will be axt.tab.
After you have a series of intensities you can run modelExp.tcl:
modelExp.tcl axt.tab nlin.spec.list 0.0
nlin.spec.list should also be created by autoFit.tcl. The 0.0 tells modelExp to use the spectral noise for estimating uncertainties. This command will fit each peaks decay, and youll get graphics, fitting statistics, etc.
The main drawback of the second method is that it requires an assigned peak table in NMRDraw, which remains NMRPipes principal weakness (and I think Frank would agree). However, if you have scripts to copy assignments, this second method isnt so bad.
Im assuming you have a good grasp on how NMRPipe and its associated programs work; if in doubt, check the man pages and web documentation before posting your processing questions on the NMRPipe list.
Good luck,
Nick
--- In nmr_sparky@yahoogroups.com , Tom Goddard goddard@... wrote:
Hi Mohammad,
Sparky has no capability to treat a 3-d spectrum as a stack of 2-d
spectra. So even if you succeed at getting this in UCSF format it wont
be useful. Sparkys ability to fit relaxation data requires separate
files for each 2-d spectrum. The pipe2ucsf program was only written to
handle real 2-d, 3-d, and 4-d spectra and probably cannot understand
some aspect of the nmrPipe format of 3-d stacks of 2-d spectra.
Best to post Sparky questions to the nmr_sparky discussion group.
http://tech.groups.yahoo.com/group/nmr_sparky/
Tom
Hello
I have collected T2 relaxation experiment in a pseudo 3D format where
axes 1 and 3 correspond to proton and nitrogen and axes 2 corresponds
to different relaxation times (11 relaxation times).
During processing of the spectra with nmrpipe, I partition the
pseudo-3D in to 11 parts each corresponding to one relaxation time.
I can open the .ft2 files in nmrDraw and they look fine.
But when I try to convert them to .ucsf files using pipe2ucsf
command, it complains that the file is smaller than expected size by a
factor of 11.
Could you let me know how I can convert them to .ucsf files for Sparky.
Regards,
Mohammad.