Is there a way to elimate false postives in Small Molecule mode?

Is there a way to elimate false postives in Small Molecule mode? Allison Haase  2019-11-04

Hi All,

Sorry for the bother, but I was wondering if there any way to set a threshold for peak intensity in small molecule mode in Skylines.

My group uses Skyline for a GC-MS data set that uses a surrogate standard for quantification. This method has a pretty noisy baseline, especially in our blanks and double blanks, as the lack of filtering causes a lot of product ions to be present.

The issue we have is with the double blanks. Due to the noisy baseline, Skyline picks out a very small peak for the internal standard and for the other molecules of interest. Due to the calibration curve being normalized to the surrogate standard, this causes incredibly high concentration values to be reported for our double blanks. Currently we deal with this problem by deleting the Surrogate Standard peaks for double blanks, but we wish to automate this process and to make sure that peaks are not picked out when they are not actually present. Is there a way to set an intensity threshold for Skyline to pick peaks?


Nick Shulman responded:  2019-11-05

There are many ways that you could exclude the double blanks from your results. You could remove those files from your Skyline document with Edit > Manage > Results, or not extract chromatograms from them in the first place.

You can tell Skyline that these files are double blanks by setting the "Sample Type" value in the Document Grid on the Replicates report.
You might want to then filter those rows out of whatever reports you are using to look at results.
If you want to learn more about the reporting and filtering capabilities of Skyline you could take a look at the custom reports tutorial:

It sounds like the problem might be that these double blanks show up on things like the Peak Area Replicate Comparison graph, and because the scale of the number is so different than the other results, it makes the graph hard to use.

It might help if you send us your Skyline document, and we could see if there's anything better we could do.

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. Otherwise, you can upload it here:

Skyline always chooses the best peak that it can find in the chromatogram. If you have a document containing peptides, then you can train an mProphet model, and tell Skyline to only integrate high quality peaks. Unfortunately, this functionality is not available for small molecule documents. The main reason for this is that we have not been able to come up with what the equivalent of decoy peptides would be for small molecules.
-- Nick