Skyline integrates low signal peak but exports a zero value

Skyline integrates low signal peak but exports a zero value halwaseem  2023-02-02 09:03


I am using Skyline Daily v to integrate small molecules using a transition list. I noticed for a few metabolites, that I manually integrated, the exported Skyline report shows zero values even though there is a peak that has been integrated. In all cases these are low signal peaks. However, within the same dataset, Skyline allows integration of noise from blanks and samples that have even lower than the biological replicates that was "zeroed out".

In the retention time comparison graph, there is a retention time window showing the sample that has a zero value. The peak area graph doesn't have a line/bar for that sample because Skyline is recording it as a zero. Please refer to attached document.

Thank you for your assistance. Let me know if there are any other details you'd like from me.


Nick Shulman responded:  2023-02-02 09:25
One thing that might be happening is that Skyline has decided that the entire signal that you are seeing is "background".
For Skyline, the background level is the lowest of the chromatogram intensity at the start and end of the peak boundaries.
Here is a page which shows how Skyline decides on the background level:

Other than that, I am not sure what could be going on. If you send us your Skyline document we can take a look.

In Skyline you can use the menu item:
File > Share
to create a .zip file containing your Skyline document and supporting files.
If that .zip file is less than 50MB you can attach it to this support request.

You can upload larger files here:
-- Nick
halwaseem responded:  2023-02-02 10:06
Hey Nick,

I look for multiple adducts per metabolite and for this particular molecule I am looking at the [M+H] and [M-H] adduct. Based on what you said about background, there is no other intensity within in or near my peak boundaries. However, I did find a tiny signal in the [M-H] feature that was present at the beginning of the chromatogram. This was not present in other samples that had equally as low signals but were NOT zeroed.

The peak that I wanted to integrate was the [M+H] adduct but somehow it seems that Skyline considers all adducts, regardless of polarity, when determining the background for a particular metabolite. Is this correct?

I deleted the [M-H] adduct from consideration and re-imported the file. This time Skyline integrated the peak in question and reported a peak area. When I put the M-H adduct back in and re-import, the M+H peak for that samples again is zeroed out.

I have attached screenshots of this. I can also provide zipped file if that is necessary.

Nick Shulman responded:  2023-02-02 10:13
It is hard to tell exactly what is going on with screenshots so, yes, please send us your that you create using the "File > Share" menu item.
-- Nick
halwaseem responded:  2023-02-02 11:00
Zipped file loaded here
Nick Shulman responded:  2023-02-02 12:06
Thank you for sending those files.

When Skyline calculates peak areas, Skyline always interpolates the chromatograms in the time dimension so that the times are evenly spaced out.
You can see what that interpolated chromatogram looks like by right-clicking on the chromatogram graph and choosing "Transform > Interpolated".
The interpolation algorithm that Skyline uses sometimes produces unexpected results depending on exactly where the newly interpolated points happen to appear on the x-axis.
In your case, you can see that whereas the raw chromatogram has one point with a nonzero intensity, when you switch the view to "Interpolated", that point gets squashed down to the x-axis.

I have put forth a proposal to change the interpolation algorithm here:

In general, Skyline is not the correct tool for integrating chromatogram peaks that have only one non-zero point in them.
-- Nick