does skyline use MS manufacturer's proprietary centroiding algorithm ?

does skyline use MS manufacturer's proprietary centroiding algorithm ? Antoine  2018-03-21

I tried to extract some Thermo QExactive data acquired in profile mode (DIA) with Skyline using the centroid mode.
It seems that skyline convert the profile data into centroid data before extraction. Is it correct?
In fact, MS data looks centroided when open with the "full scan" window.

If it is correct, does skyline use the Thermo's algorithm for centroiding the profile data?
That would be very convenient.

Thank you

Brendan MacLean responded:  2018-03-21
All correct. Yes. Skyline uses vendor centroiding and never a custom algorithm. If you want to use a custom algorithm then you need to convert to a standard format like mzML or mzXML and import that.

Antoine responded:  2018-03-21
fantastic :)
Brian Pratt responded:  2018-03-21
Hi Antoine,

That's correct: when you ask for centroiding, the scan data presented to Skyline by the ProteoWizard library layer is centroided using vendor-proprietary algorithms (when available - otherwise ProteoWizard uses its own, just as with msconvert).

Thanks for using the Skyline support board!

Brian Pratt
Brian Pratt responded:  2018-03-21
Correction to my late-arriving answer: When centroiding is desired but vendor-proprietary centroiding isn't available, Skyline actually refuses to proceed rather than falling back onto the ProteoWizard algorithm. As Brendan says, that non-proprietary option is available but only by way of conversion to mzML before using with Skyline.

Of course this has no bearing on your case, or indeed on most cases as most vendors make this available now. But I thought I should clarify.

Tobi responded:  2019-02-15
Hi all,

just for whoever changes from profile to centroid in skyline might check if the spectra actually get centroided.

I want to push this post since i experienced sth. which might have been a light bug in the current skyline daily 4.2. I had MS2 Profile data from q exactive HF. I saved the skyline document into a new seperate file, changed MS/MS filtering from Targeted/Orbitrap to Targeted/centroided 5ppm. I reimported Thermo raw files by Manage Results, the scan isolation width was reduced to 5ppm but the spectra were not centroided. The issue was solved by removing raw files completely; saving, closing and reopening the skyline document and making a new import of the raw files leading to 5ppm extraction of centroided data from profile raw files.

In principle not a big deal which would require urgent fixing, but I hope it prevent errors and helps troubleshooting.
Nick Shulman responded:  2019-02-15

It was recently pointed out that if you do a reimport after changing to or from "centroided" in your Transition Full Scan settings, the reimported file does not use the new value of that setting.
Here was the support request:

I will try to fix this bug. I believe the bug has existed since 2015, and nobody noticed it before because not many people were changing their centroiding settings between reimports.
-- Nick