How does the MS1 annotation work?

[Adafede]

Original question posted by @adrien-jagora in #26:

I had one additionnal question regarding the optional MS1-based annotation on Lotus. If I understand correctly, this happens when the score S1 equals 0 as a way to allow TimaR to do its taxonomical annotation later on. Is this MS1 dereplication based on the exact mass of the precursor ion considered or on a molecular formula previously generated ? Thanks in advance !

[Adafede]

Answer:

The MS1 annotation tries to annotate all features, independently of their S1 score. If a feature has an MS2 annotation, the MS1 annotation part will still be performed, but its result will have a spectral score of 0 (logically).

This allows to compensate for multiple shortcomings at once:

• Annotate features without fragmentation spectrum
• Annotate features with fragmentation spectrum but no library match, based on exact mass match
• Annotate weird, rarely occurring adducts/losses/dimers/multicharged

The MS1 annotation is performed as follows:

Depending on the ionization mode, [M+H]+ and [M-H]- are always considered. Then, within a relatively small retention time window, the differences between the precursor masses of all features within that window are computed.
Example:
You have 10 features between 1.25 and 1.30 min, it calculated the mass difference between all of them and then tries to match it to list of known adducts,clusters, or in source fragments or dimers.

The underlying lists are available at inst/extdata.

These are pre-combined with lists of known masses/molecular formulas of interest, by default the ones in inst/params/default/prepare_libraries_sop_merged.yaml and inst/params/default/prepare_libraries_adducts.yaml.

You can see the default parameters for the MS1 annotation at inst/params/default/annotate_masses.yaml

I hope this clarifies. If not, do not hesitate and I'll try developing further. 😉

[adrien-jagora]

Well, I think I have all the information I need ! Thank you very much Adriano !