Peak removal from multi-injection replicate Julienlab  2020-10-06 13:40
 

Hi,

I am analyzing PRM data collected as a multi-injection replicate (2 injections per sample) however for ~10% of my targets, skyline is quantifying the same peptide in both injections when it should only be quantified in one.

I have attached a pdf document showing an example of this where the first page is the correctly identified peak from the second injection as found in all four samples. The second page shows the chromatogram from injection 1 which should not identify this peptide, however it is still being quantified. The dotp scores in these cases are low and appear to be a combination of the two injections instead of only injection 2.

Is there any way to remove single injections from one peptide so the dotp score only reflects one injection?

Thank you,
Bridgette

 
 
Nick Shulman responded:  2020-10-06 13:52
When you have multiple result files per replicate, Skyline figures out which result file has the most transitions with a chromatogram.

In a PRM experiment, a MS2 transition will have a chromatogram if there is at least one MS2 scan in the result file which matched the precursor.
In your file, it appears that Skyline has decided that there are MS2 scans in both of the injections that match your precursor.

What setting do you have for the MS/MS Filtering Acquisition Method at Settings > Peptide Settings > Full Scan? If you have that set to "Targeted", then Skyline will match an MS2 scan to the one peptide in your document whose m/z most closely matched the isolation window of the spectrum.

If you are seeing this problem with that set to "Targeted", then I think maybe your method is incorrect, and you really were telling the mass spectrometer to collect data for that peptide in both injections.

If you have that set to "DIA", then it might be that you need to narrow what Skyline thinks the isolation window is so that Skyline does not think that spectra can be used by peptides that you did not intend them to be.

If you would like, you could send us your Skyline document and raw files.

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.

You can upload that .zip file and your .raw files here:
https://skyline.ms/files.url
-- Nick
 
Julienlab responded:  2020-10-06 14:56
Thanks for the response.

I do have the MS/MS filtering set to "Targeted".

I have uploaded my skyline document (20200919-skylineprm.sky.zip) and raw files of one of the replicates with both injections (KO1-1 and KO1-2).

Thank you,
Bridgette
 
Nick Shulman responded:  2020-10-06 16:38
Thanks for sending those files.

In the .raw files you sent me, the peptide L.HIVERPYSGFPDASSEGPEPTQGEAR.A has matching spectra in both of those .raw files.

That peptide has a precursor m/z of:
725.3468

In the file "KO1-1-upload.raw" there are some spectra whose precursor m/z is 725.3468 which apparently are supposed to match that peptide.
In the file "KO1-2-upload.raw" there are some spectra whose precursor m/z is 725.3905 which you probably did not want Skyline to match.

There are two ways that you could tell Skyline that 725.3905 is not supposed to match with 725.3468.

One way is to go to:
Settings > Transition Settings > Instrument
and change "Method match tolerance m/z" to a lower number. Currently it is set to the default value which is .055. You set that to .0001.

The other thing that you could do is add to your Skyline document a peptide whose m/z is 725.3905. When your ms/ms acquisition method is "Targeted", Skyline will only match spectra to the closest matching peptide in your document.

-- Nick
 
Julienlab responded:  2020-10-07 14:22
Hi Nick,

Changing the method match tolerance did not work. When I narrow the match tolerance, both spectra are removed.

Adding peptides with the precursor does work for some peptides but not all. For example, the peptide  C.VVVENSTGSR.K with an m/z of 566.8015 should only show in "KO1-2-upload.raw" and adding in peptides with similar masses from  "KO1-1-upload.raw" of 565.88741m/z and 566.621293 m/z did not remove the assignment from "KO1-1-upload.raw". I cannot see any other possible masses that would be miss-assigned to this peptide.

I am able to manually assess the data, but it would be easiest if it were possible to remove one of the matched spectra individually.

Thank you,
-Bridgette
 
Nick Shulman responded:  2020-10-07 14:45
When I changed the Method Match Tolerance to 0.0001 and imported those two raw files into a single replicate, the peptide VVVENSTGSR only had chromatograms displayed for one of the two files.
I have attached my .sky.zip.

Can you send me your .sky.zip where it appears to have not worked?
-- Nick
 
Julienlab responded:  2020-10-08 08:19
I have uploaded the file as 20200919-V2.

Thanks,
Bridgette
 
Julienlab responded:  2020-10-08 08:33
I just re-imported all of my raw data and it is working now. Before I was re-scoring the results instead of re-importing.

Thank you for your help!

-Bridgette