search

INRA / Rnmr1D

Rnmr1D is the main module in the NMRProcFlow web application
http://nmrprocflow.org/
8 stars 1 forks source link

No proper ppm scaling at different NMR-Machines #3

Open mkhpfister opened 2 years ago

mkhpfister commented 2 years ago

Hi Daniel, we ran into a problem when reading in data from two different NMR spectrometers. Both are Bruker instruments, one of them with a NEO console. The problem is that the TSP signal is not zeroed correctly. Depending on which spectrometer is first in the listing, the spectra of the second spectrometer are shifted to it. The shift is about 0.04 ppm. It seems that the TSP signal is not detected correctly. In the macro, only uniform bucketing is performed and no further processing steps are made. Are we making a mistake in the reading?

With best regards Michael

djacob65 commented 2 years ago

Hi Michael,

Sorry for may answer delay.

NMRProcFlow has been developed especially for Metabolomics approaches. So, we assumes that all spectra are acquired in the same conditions (instrument, sequence, solvent, buffer, pH, ppm range, ).

Nevertheless, we also could compare different conditions in the same spectra set. Indeed, in this latter case, some issues can occurs.

Is it possible to send me a very small subset of your spectra so that I can test and in this way could improve the software ?

Regards Daniel

mkhpfister commented 2 years ago

Hi Daniel, I just sent you an email from my work with some spectra. I hope it is helpful; otherwise, get back to me if more spectra are needed. Warm greetings Michael

djacob65 commented 2 years ago

Hi Michael,

By using the 1r spectra, it assumes that the PPM calibration was made before within TopSpin.

As it is not the case, we could add as the first macro-command :

calibration -0.024 0.074 0 10.2 10.5

This would do the job.

In addition I saw that you have two spectra sets very different (to say the least). I needed to normalize (CSN) the whole set within NMRProcFlow to see the second spectra set. There are a factor of 10^5 between them! I suppose you want to compare these two wine sets.

Best wishes Daniel

mkhpfister commented 2 years ago

Hi Daniel, thank you very much for your support. We were able to successfully read the spectra with your suggested changes in the macro. However, we had to adjust the parameters a bit to read in our large dataset (> 400 spectra, four spectrometers).

I have one comment though, when we look at the 1r spectra in TopSpin, the TSP peak is correctly set to zero for all spectra. If we only read in the ANVANCE NEO console spectra with Rnmr1D, the TSP is also correctly set to zero. The problem arises only if we read in both types of spectra together with Rnmr1D. In this case, the spectra that are first in the order (the order of spectra in the folder from which is read in) are set to zero correctly in each case. The others are shifted accordingly. That seems strange to me. I have not yet been able to find out the reason for this.

And yes, you are right, comparing these spectra is a big challenge that we are still working on at the moment.

Many greetings Michael