MZ matching is confusing in mass.csv.file when there are two fields mztheo and mzexp.

ceabiodb / biodb

An R package framework for accessing biological and chemical databases and developing or extending new connectors.

GNU Affero General Public License v3.0

11 stars 2 forks source link

Closed pkrog closed 3 years ago

pkrog commented 3 years ago

When defining a mass.csv.file, if there are both columns peak.mzexp and peak.mztheo, which one is picked by biodb for matching?

[x] Emit a warning when a match is run and both values exist, stating which one is currently used.
[x] Implement a method inside mass.csv.file to set which field to use for matching.
[x] Implement a method to get which field has been chosen.
[ ] Better : if no field is chosen when running match, choose mzexp by default and emit a warning. Or do it at loading time and emit warning.

pkrog commented 3 years ago

The ambiguity may arise also for a remote database defining both fields for its spectra. Hence it must be solved :

at BiodbMassConn level
having some default configuration (like the order of priority between fields: c("peak.mzexp", "peak.mztheo")).
storing configuration for the targeted connector only.

pkrog commented 3 years ago

BiodbMassdbConn functions getMatchingMzFields() that returns the fields to use (in priority order) inside matching functions and setMatchingMzFields() to set them.

pkrog commented 3 years ago

[x] Write tests for matching with both fields peak.mztheo and peak.mzexp in database.
[x] Update mass.csv.file vignette.