NMR Phasing doest not seem to work

[rsalek]

Hello,

I'm trying your web app ATM, with the examples given. Seems auto-phasing doest not work and keep asking to use in in time domain. Which doest not able to phase either, as far as I can understand phasing can be done in frequency domain after FTT of the FID. Any suggestion?

Reza

[ahmohamed]

Hi Reza,

Phasing an NMR spectrum requires both the absorptive (real part) and dispersive (imaginary part) spectra to be present (i.e. the spectrum has to be complex). However, when the spectrum is obtained from processed data (pdata in Bruker and *.ft in nmrpipe), only the absorptive spectrum is available, and so, phasing can no longer be done. This is the case with the adenosine examples (sample set spectra).

I agree that the message is not exactly correct. Generally, I check whether the original spectrum was in the time domain, because then it is guaranteed that both parts are available. If not, I display an error that the original spectrum has to be in the time domain.

Also, I currently apply phasing to the time-domain spectra, because that is the code I have. I suppose it is possible to apply phasing to the frequency domain (again if it is complex), but I don't have the code for it. If you can help me out, I can incorporate it in the tool and fix the error message.

Let me know if you have any suggestions.

Ahmed.

[rsalek]

Hi Ahmed,

I have local version installed and seems working nicely. I was planing to try it with some of my datasets. I was thinking of adding it as pat of Galaxy workflow for NMR data processing, any thoughts about it? I will looking into phasing.

Best Wishes
Reza

On 9 Jun 2016, at 08:52, Ahmed Mohamed notifications@github.com wrote:

Hi Reza,

Phasing an NMR spectrum requires both the absorptive (real part) and dispersive (imaginary part) spectra to be present (i.e. the spectrum has to be complex). However, when the spectrum is obtained from processed data (pdata in Bruker and *.ft in nmrpipe), only the absorptive spectrum is available, and so, phasing can no longer be done. This is the case with the adenosine examples (sample set spectra).

I agree that the message is not exactly correct. Generally, I check whether the original spectrum was in the time domain, because then it is guaranteed that both parts are available. If not, I display an error that the original spectrum has to be in the time domain.

Also, I currently apply phasing to the time-domain spectra, because that is the code I have. I suppose it is possible to apply phasing to the frequency domain (again if it is complex), but I don't have the code for it. If you can help me out, I can incorporate it in the tool and fix the error message.

Let me know if you have any suggestions.

Ahmed.

—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub #2 (comment), or mute the thread https://github.com/notifications/unsubscribe/ABUk9LyGzujR6rfEeGtw5YU9QSZjfDjAks5qJ8ZWgaJpZM4Iw3fx.

[ahmohamed]

Hi Reza,

I thought about Galaxy when I started this project, but I didn't go for it because Galaxy didn't allow spontaneous interactivity. I wanted the web component to have the look and feel of desktop NMR processing software (ACD labs, Mnova, ..., etc).

It is possible to integrated NMRPro in a Galaxy workflow for batch processing of metabolomics datasets. If I were you, I would use nmrpro python package to provide the dataset processing functionalities (see point #5 in the tutorial), and use specdraw.js in interactive-view mode for visualization.

There is a bit of work that needs to be done to make the tool config file. If you are interested in collaborating on that project, just let me know and I would be happy to help.

Regards.
Ahmed.

[rsalek]

This is great, will get in touch soon :)
Reza

[rsalek]

How are we doing