search

HUPO-PSI / mzTab

mzTab Reporting MS-based Proteomics and Metabolomics Results
https://hupo-psi.github.io/mzTab
37 stars 16 forks source link

Reporting stable isotoped labeled metabolites #179

Open MKoesters opened 4 years ago

MKoesters commented 4 years ago

Hi all,

I'm trying to implement a mztab writer for our software, however I need to be able to report metabolites labeled with stable isotopes using different enrichment levels. From reading the documentation, I could not figure out if this is possible and if yes, how. I only found this https://github.com/HUPO-PSI/mzTab/issues/90#issuecomment-367665429 comment, suggesting its not supported in mztab 1.1-M. Is this also true for other mztab versions?

I would appreciate if you could tell me how to do it or point me into the right direction.

Best, Manuel

nilshoffmann commented 4 years ago

The small molecule feature section supports to report the isotopomer: https://github.com/HUPO-PSI/mzTab/blob/master/specification_document-releases/2_0-Metabolomics-Release/mzTab_format_specification_2_0-M_release.adoc#645-isotopomer

Thus, if I understand your use-case correctly, you would have multiple SMF rows, one for each enrichment level. The question is, whether you would report them as the (monoisotopic) representative in the summary level (SML), referencing multiple SMF IDs (SMF_ID_REFS) or also as separate rows to be able to disambiguate them, then only referencing a single SMF ID per row.

You could also consider reporting the adduct_ions column (on SML level) to reflect the labeling, e.g. following https://skyline.ms/wiki/home/software/Skyline/page.view?name=adduct_descriptions This would capture the number and nature of the isotopes present on that particular molecule. At least this is what we do for labeled lipids.

Maybe @sneumann and @andrewrobertjones agree or could add to the discussion?

nilshoffmann commented 4 years ago

About the adduction_ions: using any modifications directly following M would currently not validate. The chemical_formula could be extended to capture the number and type of labeled atoms, like H[2H]O. But I am not sure whether this is "standard" notation. Usual Hill notation has prefixed superscripts to indicate the isotope and subscripts to indicate the number of occurrence of that atom.

sneumann commented 4 years ago

Hi, some work in the area (not directly mentioning mzTab) has been presented by Phillipe Rocca Serra in https://zenodo.org/record/3258123#.XdftTtEo_5c

MKoesters commented 4 years ago

Thank you for your quick replies! I think the second option (which you also use for labeled lipids) makes sense to me and is easier to implement. Do you think it would make sense to report the chemical formula using brackets? Something as 13C(3)15N(2) could work to circumvent using super/subscripts and make it easier to read. I'll also check the resource from Phillipe, thanks. So, as far as I understood, there is no real standard, but different reasonable approaches which might become standard?

Best, Manuel

andrewrobertjones commented 4 years ago

Hi Manuel, Please you could draw up a sketch file or image showing the experimental design, and we will advise from there (sorry for slow reply!)