On Wed, 26 Jun 2019 at 10:02, Yulian Gavrilov <[email protected]> wrote:
> Thank you.
> Actually I used CCPN analysis ands saved my data in Sparky format.
> In CCPN I used simple box sum method. This should be compatible with error 
> propagation, right?

Ah, that's different then ;)  Note that you cannot use their
"truncated box sum" method.  So have a look at
https://www.nmr-relax.com/manual/spectrum_integration_points.html.  If
N is different for each peak, it might require some manual work to get
all the information in.  Why do you save in Sparky format?  Is there a
CCPN output format with the volumes and the number of points in the
box?  If so, we might be able to add relax support for such input data
and automate the process.


> However, there is another problem. I don't see the way to get base plane 
> noise rmsd from CCPN.
> I wonder if I can use Sparky for example, only to get rmsd of the base plane 
> noise and CCPN only to get the volume of the peaks.

This is more difficult due to the scaling/truncation some softwares
perform.  You could cross check that the peak height is 100% identical
in both - and only then can you combine the data.  But does the "box
sum" method in CCPN only report the sum, or does it include a standard
deviation?  I suggest contacting the CCPN people to explain the
problem and to ask them if there is a way to measure the RMSD of the
baseplane noise within a user-defined box.  Note that, due to spectral
processing and spectral artefacts, different regions of the NMR
spectra have different RMSDs.  In a HSQC, the chopped out side of
spectra, right of the water signal, almost always has a lower RMSD
than around your peaks.  From my own personal experience, contributing
factors include a baseplane roll (of the noise, not the average) due
to truncation of strong peaks, noise due to impurities in the NMR
sample, proximity to the water signal, proximity to the edge of the
spectrum, and window function choice.  So a whole spectrum RMSD
measurement is quite useless for a relaxation analysis.

Regards,

Edward


_______________________________________________
nmr-relax-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/nmr-relax-users

Reply via email to