creating custom rule table for CAMERA

sneumann / CAMERA

This is the git repository matching the Bioconductor package CAMERA: Collection of annotation related methods for mass spectrometry data

11 stars 22 forks source link

creating custom rule table for CAMERA #47

Open vlagani opened 4 years ago

vlagani commented 4 years ago

I am processing a set of metabolomics profiles using xcms+CAMERA. Particularly, and I am trying to create a custom rule table for adduct identification. Unfortunately, I have not found much material on the subject, except section 6 of the "LC-MS Peak Annotation and Identification with CAMERA" vignette. Is there any additional resource where I can better learn how to create a rule table? What baffles me is the usage of the "quasi" and "ips" fields, and the corresponding concepts of annotation group and rule score. In my understanding, setting all quasi and ips values to 1 should indicate that all candidate adducts must be considered independently and have the same weight. Is this correct?

stanstrup commented 4 years ago

You might be interested in the rule set here: https://github.com/stanstrup/commonMZ/

My understanding:

An annotation group must include at least one ion with quasi set to 1 for this adduct. It often makes sense that you have to see either M+/-H or a common adduct like Na, NH4, Cl or FA.
ips is the rule score. Only the annotation hypothesis with the largest score sum survives.

Yes, I believe as you said that setting both to 1 will make all adducts have equal weight. The annotation hypothesis with the highest number of annotated peaks would thus win no matter how obscure the combination is. That is probably not a good idea...

vlagani commented 4 years ago

Hi, at the end we decided to use CAMERA functions (i.e., "CAMERA::readLists()" and "CAMERA::generateRules()") for creating a custom set of rules starting from specific ions, neutral additions and neutral losses . Thanks @stanstrup for your kind reply.