Looking for Peak Picking Algorithm Adjustment Documentation

support
Looking for Peak Picking Algorithm Adjustment Documentation jeff  2024-08-05 07:56
 

I'm trying to use Skyline to replace Tracefinder in my workflow. I'm collecting HRMS on a Thermo Orbitrap in MS1 full scan mode. I've got a list of targeted analytes, and while Skyline in general works well, all the documentation I can find for small molecules recommends making manual adjustments to peak integrations, which is not feasible for the number of samples I need to process per day.

I haven't exhaustively read every document, so I'm hoping there is a resource I've just missed!

 
 
Nick Shulman responded:  2024-08-05 08:35
What sort of problems are you running into? Could you send us your Skyline document?
In Skyline you can use the menu item:
File > Share
to create a .zip file containing your Skyline document and supporting files including extracted chromatograms.

If that .zip is less than 50MB you can attach it to this support request.
You can always upload larger files here:
https://skyline.ms/files.url

I am supposed to make peak picking better in Skyline.
I have been working on a feature where Skyline would detect when the peak boundaries in the different replicates are different from each other, even after performing retention time alignment and then adjust the peaks in the worse-looking replicates so they all match.
There is a link to where to download that special Skyline as well as a poster from ASMS this year here:
https://skyline.ms/announcements/home/support/thread.view?rowId=65188

-- Nick
 
jeff responded:  2024-08-13 08:46
Thanks for getting back to me Nick.

There are a few different issues. The "clearest" in this dataset is Adenosine-5-Diphosphate. There is a small contaminate peak right before the primary peak (4.3 vs 4.5). Even with an explicit RT of 4.6 set, the fitting algorithm sometimes prefers the 4.3 peak (which is a better looking peak).

I'll be very interested in that feature when it's testable/implemented.
 
Nick Shulman responded:  2024-08-13 09:28
Thank you for sending that Skyline document.

I can see from the Audit Log (View > Other Grids > Audit Log) that you changed the "Explicit Retention Time" value for that molecule from 4.34 to 4.55.

If you ever have questions about why Skyline has chosen a particular peak, you should use the:
View > Other Grids > Candidate Peaks menu item.

I can see that Skyline has chosen the incorrect peak because that peak is scoring very high on the "Identified Count" score.
That is, even though you moved the Explicit Retention Time, Skyline is still using the scores that the peaks were assigned when the chromatograms were originally extracted.

You can fix this by using the "Rescore" button at "Edit > Manage Results".

If you do that, Skyline will choose the correct peak.

The "Identified Count" score is so powerful that Skyline will always choose the peak that intersects with the Explicit Retention Time no matter how bad that peak is.
-- Nick