Peak annotation in Skyline mnissa  2023-02-14 12:06
 

Dear Support Team,
I had a doubt regarding the peak annotation is Skyline.
Under "Transform" option, there are four different options: None, Interpolated, Second derivate, Savitzky-Golay Smoothing.
In each view, the number of data points shows a different value (as in the attached picture)
While selecting the peak, which view should be considered for analysing the peak width or to know the number of data points in the peak? Can we consider the Smoothing view for this as that shows the maximum number of data points.

I look forward to this.

Best Regards
Mehar

 
 
Nick Shulman responded:  2023-02-14 13:25
If you want to know the actual number of points in the extracted chromatogram, the view that you should look at is "View > Transform > None".
When you have the transform set to "None", the small vertical red dotted lines at the bottom of the graph will exactly match the vertices in the chromatogram graph.

When you select a different transform, such as "Transform > Interpolated", the vertical red dotted lines are not supposed to change. They are supposed to show you the times of the actual extracted chromatogram points, regardless of how the rest of the graph has been transformed.
I am really confused as to why the number displayed next to them is different in your screenshots.

It would be very helpful if you could 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 file is less than 50MB you can attach it to this support request.
You can upload larger files here:
https://skyline.ms/files.url

-- Nick
 
mnissa responded:  2023-02-14 15:48
Dear Nick,
Thanks for your response and clarification.
I am sorry for this confusion about changing the number of data points. The number of data points was different because I changed the "peak boundaries".
If I manually change the peak boundaries, the number of red vertical lines changes.
 If I select the peak boundaries based on “None” view, it seems as if a part of the peak is not selected (if I look in smooth view) and vice versa if I select the peak boundary in Smooth view, it doesn’t cover the full peak in the other view. Also, If I select the peak boundaries based on Smooth view, it mostly shows more number of data points than if I select the peak boundaries in “ None” view.
So, my question is how should I select the peak boundaries, Should I do it by looking in the “None” view or Smooth view. OR should I not change it manually and proceed with the ones already done by Skyline.

I have attached a screenshot explaining my query and a Skyline document (named as 20230214_Mehar_L1).
Thanks and regards
Mehar
 
Nick Shulman responded:  2023-02-14 19:20
When you choose "View > Transform > None", you are looking at the chromatogram points that Skyline extracted from the spectra. The chromatogram curves is drawn as a series of straight line segments, and every vertex between line segments is a spectrum from which a chromatogram point was extracted. If you click the mouse on one of those vertexes Skyline will display the spectrum which resulted in that point.

The algorithms that Skyline uses for peak finding and integration only work on chromatograms where the points are evenly distributed in the time dimension. For this reason, after Skyline has finished extracting chromatograms, the first thing that Skyline does is resample the chromatograms in the time dimension so that all of the points are evenly spaced. You can see what that resampled chromatogram looks like by choosing "View > Transform > Interpolated". Whenever Skyline is calculating peak areas, or peak heights, or background levels, Skyline is looking at the interpolated chromatogram.

The peak detection algorithm that Skyline uses works by first taking the second derivative of the interpolated chromatogram, and then looking for the places where that second derivative crosses zero. If you choose "View > Transform > Second Derivative", it is sometimes helpful in terms of answering the question of why Skyline did or did not detect a peak in a particular place, but is probably not helpful for anything else.

If you choose "View > Transform > Savitzky-Golay Smoothing" you can see a chromatogram which is sometimes less noisy and easier for a human to look at.

The transforms "Interpolated", "Second Derivative" and "Savitzky-Golay Smoothing" all have vertices at exactly the same time points.
If you adjust the peak boundaries by clicking and dragging on a chromatogram graph, Skyline will always make it so that the peak boundary ends up being exactly on one of the Interpolated points. That is, even if you are looking at the raw chromatogram "View > Transform > None", your chosen peak boundary will end up being exactly where on one of the points in the Interpolated chromatogram.

If you would like to learn more about this resampling and peak detection, there is some interesting stuff in section 3 of Dr Pino's "Skyline Ecosystem" paper:
https://analyticalsciencejournals.onlinelibrary.wiley.com/doi/epdf/10.1002/mas.21540

There might also be some helpful information in "How Skyline calculates background and height":
https://skyline.ms/wiki/home/software/Skyline/page.view?name=tip_peak_calc

-- Nick
 
mnissa responded:  2023-02-15 12:20
Dear Nick,
This was very helpful. Thank you so much.

Best regards
Mehar