mzml inputs for SearchGUI?

39 views
Skip to first unread message

Pratik Jagtap

unread,
Aug 10, 2015, 2:06:24 PM8/10/15
to Galaxy for Proteomics, Jim Johnson, Marc Vaudel, harald.barsnes, Timothy Griffin
Hello Marc and Vaudel,

Hello - here is a request about SearchGUI / PeptideShaker.

Downstream in our workflows, we intend to visualize spectral annotations of assigned peptides (PSMs) for selected peptides. Previously, in our workflows (for ProteinPilot searches within GalaxyP) using a tool called PSM Evaluator (PSME) we have been using peptide summary outputs (of selected peptides) and mzml files to generate Lorikeet / ProtVis outputs with which we can visualize spectral annotations for these selected peptides.

For SearchGUI/PeptideShaker, JJ intends to use mzidentml outputs and mzml inputs in order to generate PSMs for visualization. However, since the inputs for SearchGUI are MGF files - we can only visualize spectral annotations on MGF files (and not mzml files).

a) Would it be possible to have mzml as inputs for SearchGUI? 
b) Alternatively, will it be possible to convert mzml files into MGF files internally within SearchGUI - or provide a means to generate a unique identifier within a header of MGF files so that we can map back to mzml files for our PSM Evaluation?

Any suggestions would be greatly appreciated.

Regards,
Pratik

Pratik Jagtap,
Managing Director, 
Center for Mass Spectrometry and Proteomics,
43 Gortner Laboratory, 
1479 Gortner Avenue, 
St. Paul, MN 55108      

Marc Vaudel

unread,
Aug 11, 2015, 3:49:47 AM8/11/15
to Pratik Jagtap, Galaxy for Proteomics, Jim Johnson, harald.barsnes, Timothy Griffin
Dear Pratik,

Using mzML as input to SearchGUI and PeptideShaker has been on our list for a long time, but never really got the time to implement it properly. For now we went for your plan b), where we convert the mzML file internally and pass the mgf along with the results using the -data option which we implemented for Galaxy. For now I am afraid this embedded conversion is only implemented for gui operation of the tool. For this we use msconvert from the Proteowizzard package. So if you run proteowizard prior to SearchGUI you will have the same results. I will try to squeeze the integration of this in our command line in the next release.

Best regards,

Marc

Björn Grüning

unread,
Aug 11, 2015, 3:59:50 AM8/11/15
to Marc Vaudel, Pratik Jagtap, Galaxy for Proteomics, Jim Johnson, harald.barsnes, Timothy Griffin
Hi Marc and Pratik,

if this is currently not available we should handle this on our (Galaxy)
side.
We already have some msvonvert wrapper and could do this step "easily"
before passing these files to SearchGUI.

"easily" means ... we need to install msconvert (windows/linux/wine???)
but this problem we also have if SearchGUI needs msconvert, but is
probably more complicated if it's bound to SearchGUI and extends with
one big dependency.

If I have a vote we should wait for proper native mzml support in
SearchGUI and do the convert-dance on our side.

Btw. I still try to get PS and SG working on our cluster and during this
I used OpenMS FileConverter to convert mzML to MGF ... is this a
possibility?

Ciao,
Bjoern

Marc Vaudel

unread,
Aug 11, 2015, 4:04:37 AM8/11/15
to Björn Grüning, Pratik Jagtap, Galaxy for Proteomics, Jim Johnson, harald.barsnes, Timothy Griffin
Hi Björn,

I confirm that if we implement the conversion we will need an installation folder for proteowizard, and I am not sure to which extend it is portable to the Galaxy. Using the OpenMS converter is a great idea :)

Best regards,

Marc

Pratik Jagtap

unread,
Aug 11, 2015, 8:48:14 PM8/11/15
to Marc Vaudel, Björn Grüning, Galaxy for Proteomics, Jim Johnson, harald.barsnes, Timothy Griffin
Hello Marc and Bjoern,

Thanks for the inputs.

Locally, we do use the Galaxy-wrapped msconvert to convert the RAW files into mzml. Later in our workflows, we use the MGF Formatter tool that converts mzml into MGF format. 

JJ will be able to comment better on when / where the scan information is lost in the mzidentml file. He will also be able to comment on whether OpenMS FileConverter is a better option than MGF Formatter so as to keep the scan mapping information intact through the workflow.

Regards,
Pratik

Pratik Jagtap,
Managing Director, 
Center for Mass Spectrometry and Proteomics,
43 Gortner Laboratory, 
1479 Gortner Avenue, 
St. Paul, MN 55108      

Reply all
Reply to author
Forward
0 new messages