points per peak in DIA data

points per peak in DIA data danielacgranato  2021-09-21

We have been starting to optimize DIA method in Exploris 240, and we are evaluating the gain or loss in points per peak based on different isolation windows.

We would like to know:
1-) how does Skyline calculate points per peak of a peptide. Does it add the points per peak given by the MS/MS spectra and the MS1 full scan acquisition? Or does it account only for the MS/MS spectra?
2-) In addition, can you share with us the best parameters to extract the chromatograms from DIA acquired data?
Thank you very much in advance. Best, Daniela

Nick Shulman responded:  2021-09-21
1. The Points Across Peak number is on the "Transition Result" level in the Document Grid. In an SRM experiment, each transition really might have a different value for the points across the peak. In a full scan (DIA) experiment, typically all of the MS1 transitions for a particular precursor will have one value for the points across the peak, and the MS2 transitions will have a potentially different value. Whenever Skyline is telling you the points across peak number, it is always telling you the number for a particular transition.

2. The Basic DIA Tutorial has been updated earlier this month, and I believe it now uses the settings that we would recommend for DIA experiments:
-- Nick
danielacgranato responded:  2021-09-21
Thank you ver much Nick for your prompt response. We understand that the number of points across peak for a particular peptide is the same for any transition given by a DIA experiment. Still we have some doubts regarding points across peak calculation in Skyline.

1- I have attached the image of a peptide that shows 6 MS/MS IDs provided by the spectral library generated with the same raw data.

2-We would assume that the peptide contains 6 points across peak, but when we export this information from Transitions Results in Skyline it indicates 12 points per peak. Where are the other points per peak coming from? Are the MS1 transitions for this peptide being accounted for?

Best, Daniela
Nick Shulman responded:  2021-09-21
When you are interested in the points across the peak, you should right-click on the chromatogram and choose "Transform > None".
After you have chosen "Transform > None", you can right-click on the chromatogram and choose "Copy Data" and paste that into something like Microsoft Excel (or Notepad). The data that you paste in there will tell you the exact retention times where the chromatograms have points.
You can also left click on a point along the chromatogram in order to bring up the Full Scan spectrum viewer. There are forward and back buttons on the Full Scan graph which will move a dot along the chromatogram, and you can use that to look at each spectrum which contributed a data point to the chromatogram.

I cannot tell, but the chromatogram in your screenshot might be the interpolated one ("Transform > Interpolated"). When Skyline is doing peak detection, the first thing that Skyline does is interpolate (resample) the points along the chromatogram so that all of the points are evenly spaced in time. This interpolated chromatogram may have more or may have fewer points in it than the original raw chromatogram. The points across the peak number always refers to the raw chromatogram.

Why are you thinking that there should be six points across the peak in that screenshot? Is it because of the six "ID" lines? Those ID lines indicate where there was an MS2 identification in your .blib file. The location of ID's is unrelated to the points across the peak. The MS2 spectrum where the peptide was identified might not be a point in the chromatogram, since it might have been a different charge state of the peptide which was identified. Also, the peptide search engine might not have identified the peptide in any particular spectrum along the chromatogram.

Usually you would not have MS/MS IDs associated with your MS2 chromatograms, because normally, your peptide search would have been performed on DDA data, and DDA data does not make for good MS2 chromatograms, since a particular precursor does not get sampled at predictable intervals. When Skyline extracts chromatograms from DDA data, the MS2 chromatograms typically have very long straight lines across the regions of time where no matching MS2 spectrum was sampled. If you would like to extract MS2 chromatograms from DDA data, we recommend that you choose "DDA" as the acquisition method at "Settings > Transition Settings > Full Scan". When the MS2 acquisition method is "DDA", the MS2 chromatograms are plotted as dotted lines indicating that Skyline is not using the MS2 chromatograms for quantification.
-- Nick
Brendan MacLean responded:  2021-09-21
Hi Daniela,
I believe the issue is with how wide the integration boundaries are. The points across the peak number includes all points between the two integration boundaries, and the right boundary in your plot is pretty far from the majority of the peak. Visually, you probably see the peak as ending at about 14.95, but your right boundary is at 15.22.

You can also see where the true points that account for the chromatogram are located by selecting its transition so that its chromatogram is highlighted in red. Then Skyline will also show vertical dashed lines where the spectra were acquired (regardless of the transformation displayed in the plot) and it will also show the points-across-the-peak number in red to the right of the integrated range.

Hope this helps. Thanks for using Skyline in your research.

danielacgranato responded:  2021-09-24
Thank you very much for all your help!! I think now we understood it better. Best, Daniela