Issue 70: Overlapping transitions causes unpredictable results during CE optimization

issues
Status:open
Assigned To:Brendan MacLean
Type:Defect
Area:Skyline
Priority:3
Milestone:4.3
Opened:2011-07-11 13:29 by Nathan Manes
Changed:2011-07-14 13:06 by Brendan MacLean
Resolved:
Resolution:
Closed:
2011-07-11 13:29 Nathan Manes
Title»Overlapping transitions causes unpredictable results during CE optimization
Assigned To»Nathan Manes
Type»Defect
Area»Skyline
Priority»3
Milestone»1.2
Skyline uses small shifts in transition m/z values to correlate fragment ion intensity with collision energy (apparently this is unavoidable). Occasionally, two transitions will overlap in elution time, precursor ion m/z, and fragment ion m/z. This is not unusual if both transitions originate from the same precursor ion.

It's undesirable that a transition list can include overlapping transitions, but this is not really a bug. Ideally, Skyline would ask for the Q1 and Q3 FWHM in the "peptide configuration" (this is set in the instrument method for a TSQ Vantage), and would warn users about any overlapping transitions (either before or after LC-SRM). This would be complicated, especially if the overlapping transitions originated from different precursors as they may elute at different retention times.

However, during CE optimization this issue causes unpredictable results (incorrectly assigned CE energies - I can't exactly figure out what Skyline did) because there are redundant transitions in the target list. Again, this isn't exactly a bug because users should avoid overlapping transitions in the first place. However, during CE optimization transition list export, Skyline could warn users if a pair of transition originate from a single precursor and have identical precursor -> product m/z values.

I didn't know who to assign this issue to so I assigned it to myself. BTW, great product - I use it quite often.
 
 Issue - close transitions cause strange results.png

2011-07-14 13:06 Brendan MacLean
Assigned ToNathan Manes»Brendan MacLean
Milestone1.2»
Hi Nathan,
I will assign this issue to myself, pretty much the default, but I am not sure when I will get a chance to look at it. So, I am going to leave the milestone blank for now.

Thanks for taking the time to give a clear explanation and post the issue.

--Brendan