Use of 54Fe isotopic peaks

[sebschmpi]

Hey everyone, I have a short question. I would like to analyze the MS data from a molecule containing an Iron and I would also like to include the isotopic peaks of M-2 and M-1 ( which include an 54Fe and none or one 13C/15N). If I define my data file with data points at a derivative at -2 and -1 in addition to 0,1, these are not shown in the output file and it seems the calculation goes wrong since the resulting peaks are very different compared to the input peaks. Thank you very much for the help. Best wishes, Seb

[pierremillard]

Dear @sebschmpi ,

M0 is by definition the lightest isotopologue, i.e. the molecule with all atoms being the lightest isotope of the corresponding element. According to this definition, isotopologues cannot be defined with a negative index. I think you should redefine your peaks starting from M0 (54Fe), the current "M0" (56Fe?) would become M2.

If this is not clear, or you believe there is a misunderstanding, do not hesitate to provide an example data file.

Hope this helps.

Best, Pierre

[sebschmpi]

Dear Pierre,

Thank you very much for your reply and your suggestions. I think my initial comment was not very precise, please excuse the inconvenience. I have added the files for a sample compound below (converted to .txt because it was not possible to upload .dat or .tsv files to github).

As you can see in the corrected_area of the _res file all area for M2, M3, M4 etc was lost and all area is now assigned to M0, which now has a larger area than before the correction.

I hope this makes my problem more clear.

Thank you again very much for the fast reply and the help.

Best wishes, Seb

MS_data_test.log MS_data_test.txt

MS_data_test_res.txt Isotopes_test.txt Metabolites.txt Derivatives.txt

[pierremillard]

No worries, we are here to help :) Thanks for the files!

To what I see, the results look as expected: in your 13C labeling experiment, the most intense mass fraction is m2 (due to the presence of one atom of iron, with 56Fe having an abundance of ~92% and 54Fe an abundance of only ~6%). Once corrected for natural abundance of non tracer elements (so corrected for iron), the most abundant 13C-isotopologue is M0.

The "corrected area" might be a misleading term, it represents the area you should have measured if all atoms other than the tracer (so including iron, oxygen, hydrogen, etc) would have been +0. In this case, the contribution of 56Fe containing molecules would have been assigned to mass fraction m0, so the area of this peak would have been higher. The isotopologue distribution is then calculated from the corrected area.

Why do you think the calculation goes wrong?

Best wishes, Pierre

[pierremillard]

Here some information regarding the interpretation of the results for the dataset you provided:

• there is no inconsistency in the measurements (all residuum are low, <0.01).
• the strong shift from M2-4 to M0-2 after correction is consistent with the presence of iron in your molecule (most naturally abundant iron isotope is +2).
• from the isotopologue distribution, we can observe a significant incorporation of one or two 13C (with M1 at 28.7% and M2 at 15.1%), and an absence of heaviest isotopologues (with 3 or more 13C atoms). I hope this makes sense for your experiment.

[sebschmpi]

Dear Pierre,

Thank you very much for your explanation. I had misunderstood the meaning of M0 after the correction by isocor, I was expecting something like M0=0, M1=0, M2=50%, M3=29% etc. This helps a lot.

Thank you also for the analysis of the sample data, I had generated a sample file by modifying a calculated isotope distribution with an additional abundance of 1x and 2x 13C and the results fit very nicely.

If you do not mind I have two more questions, one is directly related to my data and one is more theoretical.

The theoretical question is in my test data set I supply 10 isotopologues (M0-M10). However, my molecule contains 10 carbon atoms and 1 iron so in principle, I should supply 12 isotopologues from 54Fe 12C10 to 56Fe 13C10. Or is there another misunderstanding on my side?

Related to my data, please see the attached files. In this data I used a nonlabeled tracer, in this case I would expect the isotopolgue distribution after correction to not contain any 13C labeled isotopologues. However, I see quite a large fraction of 1x13C labeled isotopologues. Did I misinterpret my results or is there another explanation? Best wishes, Sebastian

12C ctrl_res.txt