Hi Ron,
I just updated my TPP binaries and I am now getting the same error as you. It appears that MzXml2Search no longer writes its results to the current working directory, instead it writes the mgf next to the mzXML file.
In your command window please run the following command to confirm this:

cd C:\tmp

then run:
C:\Program Files\LabKey Server\bin\MzXML2Search -mgf D:\CPASdata\G056\B_090310_09_G056L9_A09.mzXML

Is the B_090310_09_G056L9_A09.mgf file in you C:\tmp directory or in D:\CPASdata\G056 ?

There is a bug in the error message.
21 May 2009 14:13:24,327 ERROR: Failed to find Mascot query file 'D:\CPASdata\G056\mascot\Mascot1\B_090310_09_G056L9_A09.work\input.xml'
Should be:
21 May 2009 14:13:24,327 ERROR: Failed to find Mascot query file 'D:\CPASdata\G056\mascot\Mascot1\B_090310_09_G056L9_A09.work\B_090310_09_G056L9_A09.mgf'

Thanks,
Bill

Sorry, the change isn't due to a change in MzXml2Search it's because my new install uses the local pipeline instead of the enterprise pipeline. This will have to fixed in the Labkey code.

Hi Ron,

I wanted to give you an update on this issue. Bill is testing a fix and once it's been fully validated I'll check it into the 9.1 branch.

Thanks,
Josh

Dear all,

I have the same problem. I have not install TPP on this server, I am using CPAS 9.1 and Mascot 2.2 on Apache HTTP server 2.2.

Thank you for your Help.

Thomas

Hi Thomas,
Due to vacation schedules the fix for this problem probably won't be available until early next, week but there is a solution in the queue.

Hi Thomas and Ron,

Thanks for your patience.

I'm attaching a patched 9.1 ms2.module that contains the fix. Please back up and then replace your existing copy. It should be in <LABKEY_ROOT>/modules. You'll then need to restart the web server and try a new search.

Please reply here with your results.

Thanks,
Josh

Hi Josh,

Thank you for your answer, however I have a new error message. I think it's a complete different problem.

Thomas

Hi Thomas,

Yes, you're getting much further now. The Mascot search ran to completions, so the patch worked. (Thanks Bill!)

The new error is that there are no peptides that were assigned a PeptideProphet probability of 0.05 or higher. Most likely all the peptides had low Mascot scores as well. There are many possible reasons for this - searching the wrong protein database, having poor quality spectra, etc.

To run the search to completion, you can add this to your analysis protocol:

<note type="input" label="pipeline prophet, min probability">0</note>

which will prevent PeptideProphet from filtering out low confidence peptides. ProteinProphet should then run successfully, and the results will be loaded into the CPAS database, where you can more easily see what the scores look like.

Thanks,
Josh

Hi Josh,

Of course Bill have made a nice job and have resolved file allocation problem. Thanks also to him.

I really have a problem with mascot pipeline. If I analyse the same file with mascot and Xtandem, I have not the same results. I am using the same setting for both search engine. Of course the algorithm is different, however it is not explain the results.

Thomas