Skip to main content

Topic: Mzmed - mz of fragment or compound? (Read 612 times) previous topic - next topic

  • ab123
  • [*]
Mzmed - mz of fragment or compound?
Hello everyone,

I'm a tad confused as to the mz XCMS spits out after pre-processing.

Are these definitely compound masses or could they be fragments (not adducts) of a compound?
I'm asking because that obviously makes a huge difference for identification and I find that my spectra for select retention times don't match the mass, i.e. they often contain base peaks that may be higher than the mzmed given by XCMS.

So are the mzmed values both compounds and fragments thereof and if it's the latter how would you account for that (if it's not an adduct)?

Many thanks!

  • Jan Stanstrup
  • [*][*][*][*][*]
  • Administrator
Re: Mzmed - mz of fragment or compound?
Reply #1
XCMS spits out "features" meaning all chromatographic peaks it can find for all masses (so think of it as XCMS doing extracted ion chromatograms for all possible masses and integrating all peaks it sees). Meaning that there might be many features in your list that come from the same compound. Some are pseudo moelcular ions, some are fragments, some are adducts, some are isotopes, some are noise, some are contaminants... You cannot even count on all compounds showing a pseudo molecular ion.

Probably the main challenge in metabolomics is exactly deciphering what is what. It is not a trivial task.
The typical first step to figuring what is what is to use the CAMERA package that try to group features according to which are likely to come from the same compound.
  • University of Copenhagen, Denmark
Blog: stanstrup.github.io

  • ab123
  • [*]
Re: Mzmed - mz of fragment or compound?
Reply #2
Thanks Jan, that does clarify it a little. Merry Xmas!
Any other suggestions greatly welcome :-)
  • Last Edit: December 20, 2017, 10:27:23 AM by ab123

  • Jan Stanstrup
  • [*][*][*][*][*]
  • Administrator
Re: Mzmed - mz of fragment or compound?
Reply #3
In 15 min at 18:00 CET there is a webinar that I think covers exactly this: https://attendee.gotowebinar.com/register/564911869539283713
  • University of Copenhagen, Denmark
Blog: stanstrup.github.io

  • Oliver Fiehn
  • [*]
Re: Mzmed - mz of fragment or compound?
Reply #4
Instead of using XCMS, try out MS-DIAL, see here
http://prime.psc.riken.jp/Metabolomics_Software/MS-DIAL/

MS-DIAL does not report features, but peak groups ('deconvolution').

  • ab123
  • [*]
Re: Mzmed - mz of fragment or compound?
Reply #5
Thank you both very much!

  • Peradiskic
  • [*]
Re: Mzmed - mz of fragment or compound?
Reply #6
Please, avoid using term pseudo molecular ion.Proper term is protonated deprotonated molecule.