There is a number of quantification procedures for 1H NMR metabolomics data. There are few main categories of this methods:

1) Integration of the signal. Any software for NMR spectra processing allows to do it. Alternatively simple MATLAB script will work just fine.

2) Fitting the NMR signal using some deconvolution algorythm.

            - For manual peakfitting (e.g. ACD/Lab NMR processor) http://www.acdlabs.com/resources/freeware/nmr_proc/

            - For automatic global spectral deconvolution (GSD) (e.g. Mnova NMR) http://mestrelab.com/software/mnova/nmr/

            - If you are familiar with R project BATMAN is worth consideration for automatic fitting of all metabolites simultaneously http://batman.r-forge.r-project.org/

3) Fitting preexisting spectral library to the metabolic profile.

            - AMIX (by Bruker)

            - ChenomX that was already mentioned few times.

Before choosing one you should consider what is the main goal of the quantification. If it is the relative comparison between groups than probably you can just use simple integration of the resonance without overlap. If you want to report actual concentrations, than more sophisticated methods are required. Keep in mind that when using ChenomX, the spectra needs to be acquired using particular parameters and NOESY1D pulse sequence. Otherwise obtained concentrations might not be accurate: (http://www.chenomx.com/news/img/page23/ChenomxAppNote.an014.pdf) 

Hi Gaia, we use Chenomx software for metabolite quantification in our lab and have found it to be excellent for quantification of most metabolites. The protocol that we use can be found on PMC4264773

Hope this helps.

Hi, Gaia, there are a number of softwares and websites that can be used to process the spectra generated from NMR.

- http://www.metaboanalyst.ca/faces/home.xhtml

- R https://www.r-project.org/

- AMIDS  http://www.amdis.net/ 

Best regards,

Thanks a lot for your answers.

I was wondering if it is possible by using Chenomx, to calculate differences in metabolites concentration of >100 spectra at the same time.

Thanks a lot in advance.

Hi Gaia,

As far as I know Chenomx sells different licence versions and on some it definitely is possible to 'batch fit' metabolites for multiple spectra. For some metabolites e.g. lactate, the computer programme is very good at automatically quantifying but for most you get better results if you do it manually. I recently did this for >250 spectra and it took a few weeks.

Batch processing can also be done using Matlab scripts, e.g. import spectra, correct baseline, integrate target peaks, calculate concentration (using PULCON method), export results to LIMS system

It depends on the type of your gaol; targeted metabolomics or untargeted metabolomics. If you would like to study the differences between concentration of special metabolites, Chenomx Software will be OK but to find out metabolites that are different between your samples; you can use binning methods such as uniform binning or AI binning.

Best Regards

I hope until you have the answers. I used the trial version of chenomx software to quantify major metabolites present in urine. Also, you can use the manual integration method to relatively quantify the metabolites with respect to the concentration of TMS or whatever internal standard you are using.

There is a number of quantification procedures for 1H NMR metabolomics data. There are few main categories of this methods:

1) Integration of the signal. Any software for NMR spectra processing allows to do it. Alternatively simple MATLAB script will work just fine.

2) Fitting the NMR signal using some deconvolution algorythm.

            - For manual peakfitting (e.g. ACD/Lab NMR processor) http://www.acdlabs.com/resources/freeware/nmr_proc/

            - For automatic global spectral deconvolution (GSD) (e.g. Mnova NMR) http://mestrelab.com/software/mnova/nmr/

            - If you are familiar with R project BATMAN is worth consideration for automatic fitting of all metabolites simultaneously http://batman.r-forge.r-project.org/

3) Fitting preexisting spectral library to the metabolic profile.

            - AMIX (by Bruker)

            - ChenomX that was already mentioned few times.

Before choosing one you should consider what is the main goal of the quantification. If it is the relative comparison between groups than probably you can just use simple integration of the resonance without overlap. If you want to report actual concentrations, than more sophisticated methods are required. Keep in mind that when using ChenomX, the spectra needs to be acquired using particular parameters and NOESY1D pulse sequence. Otherwise obtained concentrations might not be accurate: (http://www.chenomx.com/news/img/page23/ChenomxAppNote.an014.pdf) 

Hi,

we use AMIX (Bruker) and the freely available software MetaboQuant. Check out our review and publications:

Article Current Experimental, Bioinformatic and Statistical Methods ...

Article MetaboQuant: a tool combining individual peak calibration an...

Hi, Gaia

You can use BAYESIL which is free, you need to prepare your samples as recommended by them. It was developed in Wishart lab, find attached document for the publication.

Article Accurate, Fully-Automated NMR Spectral Profiling for Metabolomics

Hi Gaia,

Have a look at http://nmrprocflow.org/ I think It might suit perfectly for what you want to do.

Hope this helps.

Best regards,

Daniel Jacob

Hi Gaia

I suggest you use Chenomx software(http://www.chenomx.com/) to quantitate the metabolites. It will be useful to determine the concentration of metabolites for biofluids. For data analysing, there are several free software available, such as Metaboanalyst(http://www.metaboanalyst.ca/).    

Best, 

Fariba 

Thank you all very much

Interesting!