mzIdentML import from PeptideShaker

support
mzIdentML import from PeptideShaker Harald Barsnes  2015-01-19 09:56
 
Hello,

When importing mzIdentML files exported from PeptideShaker (http://peptideshaker.googlecode.com) to use as spectrum libraries in Skyline we get an error about an invalid CV parameter (see attachment). However, I'm not able to figure out which CV parameter this error is referring to? Is there a Skyline log file somewhere with more details?

Attached is also the mzid file in question (the example dataset from PeptideShaker).

If you can help us figure out the source of the issue we should be able to make the required changes to our mzid export and thus make it compatible with Skyline.

Thanks for the help.

Best regards,
Harald

(PeptideShaker developer)
 
 
Nick Shulman responded:  2015-01-19 10:54
Skyline uses another program called "BlibBuild.exe" to construct the spectral library.

I ran BlibBuild.exe under a debugger, and I can tell you that the thing that it is complaining about in your file is where it says "MS:100XXX".

I hope this helps. I don't know whether this is something that we need to fix about BlibBuild (certainly the error message could be made more helpful).
 
Harald Barsnes responded:  2015-01-19 11:37
Hi Nick,

Thanks for the help. "MS:100XXX" was something we used as a placeholder while working on the new mzid 1.2 format changes, but should of course have been removed.

After removing the incorrect CV term I'm able to start loading the file, but I now get the following error message instead: "ERROR: Can't determine cutoff score, unknown analysis type".

Any tips on this one as well?

Best regards,
Harald
 
Kaipo Tamura responded:  2015-01-19 11:57
Hi Harald,

When building libraries, BlibBuild first tries to identify the software that produced the mzIdentML file in order to determine how to do further processing (e.g. what scores to look for). We will need to add support for this case.

Thanks,
Kaipo
 
Harald Barsnes responded:  2015-01-19 13:07
Hi Kaipo,

Thanks for the information. I'm attaching the revised mzid file from PeptideShaker in case you can extract what you need from this file.

In case you don't know PeptideShaker (http://peptideshaker.googlecode.com), it is a tool for the interpretation of proteomics identification results from multiple search engines. At the moment we support X!Tandem, MS-GF+, MS Amanda, MyriMatch, Comet and OMSSA, and more are on the way. All of these search engines are supported via SearchGUI (http://searchgui.googlecode.com), which is where we mainly take our input from.

After processing the individual search results these are combined into a single PeptideShaker result. Both the combined PeptideShaker result and the individual search result scores are included in the mzid file.

Please let me know if you need any additional information about either PeptideShaker or the content of the mzid file.

Best regards,
Harald
 
GGIL responded:  2015-02-02 08:43
Hi Kaipo and Harald,


Will new coming updated version support mzid file from PeptideShaker? or is there any test release from Skyline Daily? PeptideShaker also can export pride.xml that is a file format Skyline supports. When I tested it, it seems Skyline doesn't support it. Any suggestion or help would be appreciated!


Gil
 
Kaipo Tamura responded:  2015-02-02 14:40
Hi Gil,

Support for PeptideShaker outputs will probably not make it into the upcoming version, but hopefully the version after this.

Thanks,
Kaipo
 
GGIL responded:  2015-02-03 09:11
Hi Kaipo,

Thanks. Hope we have the support soon.

Gil
 
phains responded:  2015-02-11 16:51
I'd just like to add my name to the people looking forward to this implementation. I use PeptideShaker all the time and would love to see it supported in Skyline, which is also one of my favourite programs...