Re: [proteowizard-support] BlibBuild question

[Mike T. <mt...@cg...>]

Thanks for the quick reply.  Yeah, it worked fine with my mzml test and not when I tried mgf, I was just hoping there might be some secret flag or something I could give it, as it would simplify things for me to stick with mgf.  I'm painfully aware of how the scan numbers get written arbitrarily by every "flavor" of mgf, and have managed to write regexs that can strip out scan numbers from at least a few different mgf headers, but yes, its kind of ugly.  

Mike



> On Mar 29, 2019, at 2:37 PM, Chambers, Matthew <mat...@gm...> wrote:
> 
> Hi Mike,
> 
> MGF is currently not supported when importing pepXML results. MzML would be best because its nativeIDs allow spectrum matching to be unambiguous. But pepXML doesn't have a standard attribute to refer to nativeID, so BiblioSpec recognizes the non-standard "spectrumNativeID" attribute on the <spectrum_query> element. Without that, it will use the "start_scan" scan number value to lookup spectra (i.e. it look for that number in a spectrum id in a mzML or mzXML file).
> 
> There's no such (reliable) way to look up scan numbers in MGF files because the scan number attributes are optional and/or mixed in with the title string.
> 
> Hope this helps,
> -Matt
> 
> 
> On 3/29/2019 5:14 PM, Mike Trnka wrote:
>> I'm not sure where best to direct Bibliospec questions, but will try here.  I am setting up a workflow that involves building blibs from protein prospector generated pep.xml results files.  Is it possible to use mgf as the peaklist format, or do I need to generate mzml?  Best,
>> 
>> Mike Trnka
>> UCSF
>