Issue 689: File > Export > Spectral Library losing information from small molecule documents

issues
Status:open
Assigned To:Brian Pratt
Type:Defect
Area:Skyline
Priority:2
Milestone:4.3
Opened:2019-10-08 17:47 by Brendan MacLean
Changed:2022-03-30 16:51 by Brendan MacLean
Resolved:
Resolution:
Related:880
Closed:
Related Issues
2019-10-08 17:47 Brendan MacLean
Title»File > Export > Spectral Library losing information from small molecule documents
Assigned ToGuest»Brian Pratt
Notify»kikirkwo
Type»Defect
Area»Skyline
Priority»2
Milestone»19.2
Kaylie has supplied us with a document where File > Export > Spectral library currently loses the following information:
1. Not all molecules are exported. Specifically, several iRT standards don't get included in the library. Though we felt condfident they have enough data in the file to be included. If they are being dropped due to some missing information or error, maybe we should present an error list to the user to make it more obvious why we are dropping targets in the document from the exported library.
2. Not all peaks in the resulting spectra are annotated when we used the "Add All" button to add all precursors from the library as targets to a new document and looked at them in the Library Match view. Probably also just in the Spectral Library Explorer. But Add All also did not add back all of the transitions (e.g. a peptide with 6 transitions might end up with 3 despite all of them having signal - peak area - in the original document)
3. Ion annotations do not preserve the ion names from the document but instead always use "ion[mass]" for the annotations when the document gave them names.

2019-12-31 15:08 Brian Pratt
Item 3 corrected in commit 16aa704b286652c81d62d24ac73452e728e59a66 for 19.2 and cherry-picked to master

Items 1 and 2 need more investigation.

2022-02-16 14:59 Brian Pratt
Milestone19.2»
Investigating this further, it looks to me like the missing things aren't going into the library because we intentionally don't export precursors that don't have fragment transitions, or precursors without chromatograms.

If that's as it should be, then I guess the work here is just to issue a report showing why things were omitted.
 
 BTLE Negative LC Ramp.7z

2022-02-16 15:00 Brian Pratt
Also I can't repro the effect described in part 2.

2022-03-30 14:39 Brian Pratt
See also "Issue 880: SSL format changes for Feature Detection" where we have decided that BlibBuild should pass through no-fragment precursors.

2022-03-30 14:49 Brendan MacLean
Related»880