Non-unique identifiers for ions about tools-metabolomics

[lecorguille]

Are we agree that it's important during the peaklist export (currently camera.annotate, camera.combinexannot and soon xcms.fillpeaks) ?

If so, do you want that those redundant ion identifiers to be spotted and corrected automatically by increasing the number of digit?

[lecorguille]

Crossref: workflow4metabolomics/camera.archive#11

[lecorguille]

Done for FillPeaks-2.0.8: https://github.com/workflow4metabolomics/xcms/blob/bfb80b96ab0497a8d1fd853b779a26897f765929/galaxy/xcms_fillpeaks/abims_xcms_fillPeaks.xml#L45

[yguitton]

just a note they're is a function called xcmss:::groupnames in xcms that can be used (with function(xcmsSet object, mzdec = 0, rtdec = 0, template = NULL) with that even if a duplicate name is found it will be converted to MxTy_z

[lecorguille]

@yguitton Good suggestion but ... how do you use groupnames

I try without template

ids <- groupnames(xset, mzdec = numDigitsMZ, rtdec = numDigitsRT, template = NULL)

but:

|  Error in strsplit(template[1], "[T_]") : non-character argument
|  Calls: getPeaklistW4M -> groupnames -> groupnames -> .local -> strsplit

Help is welcome because they isn't so much documentation 👎

[lecorguille]

@melpetera
Is make.unique is enough? here

ids <- make.unique(paste0("M",round(groups[,1],numDigitsMZ),"T",round(groups[,4],numDigitsRT)))

[lecorguille]

@melpetera @yguitton UP!

[melpetera]

About make.unique: I would add the sep="_" argument because identifiers already have points in it.
But I think this can do!

[lecorguille]

From xcms.group 2.1.1 and xcms.fillpeaks 2.1.1
as discussed here: #36

[lecorguille]

docker run -d -p 8080:80 quay.io/workflow4metabolomics/galaxy-workflow4metabolomics:beta

[lecorguille]

Available within the dev galaxy instance

[melpetera]

Tested on dev galaxy instance!
It is ok for me.

[lecorguille]

• validated by @melpetera
• validated by @yguitton
• validated by @jfrancoismartin

[lecorguille]

Validated