Importing pepXML, mzid, or mzTab from OpenMS in Skyline

support
Importing pepXML, mzid, or mzTab from OpenMS in Skyline lparsons  2019-04-16
 

I am attempting to import Peptide Search results into Skyline (4.2.0.19072) that were obtained in OpenMS (in this case using the Comet search engine). The default output from OpenMS is idXML, which is not supported in Skyline. I looked at the options in File -> Import -> Peptide Search and found three possible formats for this though each returned a different error:

pepXML : "file is not from one of the recognized sources"
mzid : "No spectra found for the new library"
mzTab : "No acceptable score type found"

I checked to make sure these files are reasonable in size; they are all 10-25 MB with 34,000 - 271,000 lines. If I instead continue to work with the idXML file in OpenMS I see plenty of hits as well. If I use the same spectra (in their original raw format) and the same database but instead run it through MaxQuant, I can import the msms.txt file without a problem.

My goal is to import the Peptide Search results and then load a series of MS1-only spectra on top of the results so that peptide quantification can be done using the MS1 spectral data and mapped to the MSMS data of the search results. This works fine if I use the MaxQuant results here instead, though I'm looking to use the Comet (OpenMS) results at this time.

thank you
Lee

 
 
Nick Shulman responded:  2019-04-16
Lee,

Skyline uses a program called "BiblioSpec" to to convert peptide search results into a spectral library.
This is the list of all of the types of search results that BiblioSpec knows how to read:
https://skyline.ms/wiki/home/software/BiblioSpec/page.view?name=BlibBuild

It knows how to read Comet, but only if it's been run through percolator and you have a .perc.xml file.
When reading .pep.xml files, BiblioSpec needs a way to figure out what the q-values are, and all of the search engines have a different way of encoding that, which is why you get the "not from one of the recognized sources" error if it is from a search engine that BiblioSpec does not understand.

Do you have any sort of data that's mentioned on the list of formats on that BlibBuild web page?
-- Nick
 
lparsons responded:  2019-04-16
Nick

Thank you for the quick reply. Unfortunately the OpenMS wrapper for Percolator is not able to write the .perc.xml format. The only formats it can write that are relevant to this are "mzid" and "idXML" (it can also write "osw" which is not useful in this case). The mzid format here produces the same error I received without running Percolator; I also tried converting idXML to mzTab or pepXML and received the same errors I mentioned above as well.

thank you
Lee
 
matt.chambers42 responded:  2019-05-28
Hi Lee,

Can you upload a Comet->Percolator->mzIdentML file? It shouldn't be too hard to add support for that.

Thanks,
-Matt
 
lparsons responded:  2019-06-05
I just uploaded an mzid file to https://skyline.ms/files.url

Thank you Matt!
 
matt.chambers42 responded:  2019-06-05
Fixed in https://github.com/ProteoWizard/pwiz/pull/564

But that file is pretty buggy. All the SpectrumIdentificationItems have rank=0 (which is invalid) and some of them have no PeptideEvidenceRefs (which is invalid).