search

MetaSys-LISBP / IsoCor

IsoCor: Isotope Correction for mass spectrometry labeling experiments
https://isocor.readthedocs.io
GNU General Public License v3.0
24 stars 9 forks source link

Formula for Correction should be "de-adducted"? #41

Open jmmitc06 opened 7 months ago

jmmitc06 commented 7 months ago

My reading of the documentation suggests that I should use the molecular formula of the metabolite for the instantiation of the corrector object; however, the observed molecular entity in the mass spectrometry experiment will be the adduct of the metabolite.

The documentation states:

''' For instance, in the following situations, the formulas should include:

for MS measurements: all atoms of the detected ion

'''

So if I measure the proton adduct of glucose, should the formula for correction be "C6H12O6" or should it be "C6H13O6" or should I treat the adduct as a derivative?

I'm specifically looking at LC-MS data and only MS1 data.

pierremillard commented 7 months ago

You should use the molecular formula of the detected ion for the instantiation of the corrector object.

If no atom of the "adduct" are also the tracer element, calculation results will be the same if you provide atoms of the adduct in the metabolite formula or as derivative. If some atoms of the "adduct" are also the tracer element, they should be declared in the derivative formula.

Regarding your example where the adduct atom is "H", if you have carried out a 13C or 15N labeling experiment, I would suggest providing the formula of the adduct as metabolite formula. If you have carried out a 2H labeling experiment, the atom of the adduct (H) should be provided in the derivative.

Let us know if this is not clear.

jmmitc06 commented 7 months ago

Thanks, this makes sense.

jmmitc06 commented 7 months ago

Wanted to ask a follow up question. Does that mean that if I assume that the adduct cannot be labelled, then including its formula in the derivative field would always be safe? This is an assumption my tool makes for all analyses. So, if it is safe, this would make my code cleaner.

pierremillard commented 7 months ago

I realized this point after having replied. You are right, if the adduct cannot be labelled, including its formula in the derivative field will be safer. In fact, this will be the safer & cleaner way of providing the adduct, I would recommend always including its formula as derivative.