Skyline not correctly importing centroided Agilent data

support
Skyline not correctly importing centroided Agilent data estancliffe  2021-10-04
 

Hello Skyline Team,

Thank you for your continued development and user support for this great tool! We have been using Skyline as our primary tool for processing LC-MS data for several years. Unfortunately, we have recently encountered a problem that is effecting a seemingly random subset of our data. When we import raw data from an Agilent 6545 QTOF where both profile and centroid data were collected into Skyline with the "centroided" option selected as the precursor mass analyzer in the Full-Scan tab of the Transition settings, with some files, the MS1 spectra are lost and flat chromatograms appear for all transitions (see screenshot_centroid_1). When looking at the individual spectra within Skyline, it shows only SIM spectra (see screenshot_centroid_2). With other files acquired around the same time and with the same instrument (I included one example that worked and one that did not), this is not a problem. I have checked the raw data in other software and the centroided spectra are contained in the raw data of these problematic files. In troubleshooting this issue, I tried converting the file to mzML (with MSConvert) with CWT centroiding the data imports correctly into Skyline and the data looks fine. When I used MSConvert with vendor peak picking, the problem is not fixed. I also tried loading this file into Skyline with the "TOF" option selected as the precursor mass analyzer and 10,000 resolving power and the data loads correctly for all files. This issue seems to be isolated to centroid data. I have included two Skyline documents (one centroid, one profile), the raw data, transition list, and screenshots. Any help or advice on next steps on resolving this problem would be greatly appreciated.

Raw data and Skyline documents will be uploaded separately and linked.

Error has been reproduced on Skyline Daily (21.1.1.223) and Skyline (21.1.0.146) running on Windows 10.

Many Thanks!
-Ethan

 
 
estancliffe responded:  2021-10-04

Linked documents and data have been uploaded: Centroid_data_error_skyline.zip

 
Nick Shulman responded:  2021-10-04
Thank you for uploading those files.

When Skyline sees an MS1 spectrum where the m/z scan range is less than 500 units wide, Skyline assumes that that is a SIM spectrum. If the scan range is greater than 500 units, then Skyline calls that an ordinary MS1 spectrum. Skyline extracts one chromatogram using the SIM spectra, and a different chromatogram using the MS1 spectra. If a precursor has both an MS1 chromatogram and a SIM chromatogram, then the chromatogram that ends up getting used is the SIM chromatogram.

It appears that Skyline is getting confused, and thinks that spectrum #16286 is a SIM spectrum. When I look at that spectrum in SeeMS.exe, I see that it is supposed to have covered the m/z scan range from 50-1000, which is much more than 500, so it should have been treated as an MS1 spectrum.
I think what is happening is that when you ask for centroided data, Skyline loses the information about what m/z range the mass spectrometer acquired data for, and instead, is looking at the actual m/z values that had a non-zero intensity.
I will ask around and find out if someone can fix this.

-- Nick
 
estancliffe responded:  2021-10-05
Thank you for the explanation and looking into this, Nick. We would really prefer to be able to use the centroided data if possible. Please let me know if you find anything more out.

Thanks again!
-Ethan