support for Morpheus search algorithm in TPP

[Craig Wenger]

Is there a mechanism for adding support for a new search algorithm to the various TPP tools? A few users have requested TPP compatibility for the Morpheus search algorithm. It already outputs pepXML.

-Craig

[David]

I would be very interested in support for using Morpheus with TPP

[Dave Trudgian]

Craig, just wondering if you or anyone else is working on this? If not I was thinking I would have a go at putting in a Morpheus model for PeptideProphet towards the end of next week.

Dave Trudgian

On Saturday, November 30, 2013 4:37:30 PM UTC-6, Craig Wenger wrote:

[Eric Deutsch]

Hi Dave, at ISB we are not planning this, so please go ahead and let us know how it goes!

 

Regards,

Eric

--
You received this message because you are subscribed to the Google Groups "spctools-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to spctools-discu...@googlegroups.com.
To post to this group, send email to spctools...@googlegroups.com.
Visit this group at http://groups.google.com/group/spctools-discuss.
For more options, visit https://groups.google.com/groups/opt_out.

[David Shteynberg]

I was dabbling in this, but running into some issues with morpheus slowing my machine to a point where I was unable to do any other work so I moved to another project.  However, I am certainly still interested in this, but I know that pepXML coming out of morpheus may need some work.  If you'd be willing to generate some pepXML's from the tutorial data and work with me to correct the pepXML issues I should be able to model the results quite quickly with PeptideProphet.  Please let me know if that sounds good to you.

Cheers,

-David  

[Dave Trudgian]

Hi David,

Am happy to generate some pepXMLs from the tutorial data, but assuming you mean the stuff at:

ftp://ftp.peptideatlas.org/pub/PeptideAtlas/Repository/TPP_Demo2009/

... I'm not sure that's the best to work with for Morpheus. It's CID Ion-trap MS/MS, which is not Morpheus' target. Morpheus doesn't work as well as other algorithms on low-res MS/MS so I'm not sure if anyone would want to use it like that.

I grabbed two of the result sets from the Coon lab site - Yeast analysis with FT HCD or IT CID. The distributions of Morpheus scores are very different. The CID data doesn't even look like target and decoy matches are separable, so I think high-res MS/MS is a must. See attached.

DT

[David Shteynberg]

I was referring to this data:

http://tools.proteomecenter.org/wiki/index.php?title=TPP_Tutorial_v2

However, this is not high res MS/MS either.  This would be a good starting point.  The decoys and targets don't have to separate by score alone, PeptideProphet should be able to model the plots you show in either case. 

Cheers,

-David

[Dave Trudgian]

David,

That's the same data. I've made PepXML and the other Morpheus output available at:

https://proteomics.swmed.edu/data/morpheus_tpp/

DT

[David Shteynberg]

I was able to make a first stab attempt at it and I was successful after minor modification to the way spectrum_names are encoded in spectrum_query tags in the morpheus pepXML files.  After running a simple sed to replace the names with something that wouldn't overlap with anoth file when combined by using the each run's basename I was able to generate results.

My PeptideProphet processed analysis for these two searches can be downloaded here:

https://www.dropbox.com/s/f5r8j5yj2p2r3b8/morpheus_interact.tgz

Here are my sed commands, but I hope this could be fixed within morpheus.

 sed -i 's/controllerType=0 controllerNumber=1 scan=[0-9]*/OR20080317_S_SILAC-LH_1-1_01/g' OR20080317_S_SILAC-LH_1-1_01.pep.xml

 sed -i 's/controllerType=0 controllerNumber=1 scan=[0-9]*/OR20080320_S_SILAC-LH_1-1_11/g' OR20080320_S_SILAC-LH_1-1_11.pep.xml

Cheers,

-David

P.S.  I expect to merge this into the SVN codebase sooner than later.

[David Trudgian]

Awesome. Look forward to trying it out.

--
David Trudgian Ph.D.

---

From: spctools...@googlegroups.com [spctools...@googlegroups.com] on behalf of David Shteynberg [David.Sh...@systemsbiology.org]
Sent: Friday, February 07, 2014 2:22 PM
To: spctools-discuss
Subject: Re: [spctools-discuss] Re: support for Morpheus search algorithm in TPP

You received this message because you are subscribed to a topic in the Google Groups "spctools-discuss" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/spctools-discuss/qdmov2KGlJE/unsubscribe.
To unsubscribe from this group and all its topics, send an email to spctools-discu...@googlegroups.com.

To post to this group, send email to spctools...@googlegroups.com.
Visit this group at http://groups.google.com/group/spctools-discuss.
For more options, visit https://groups.google.com/groups/opt_out.

---

UT Southwestern Medical Center
The future of medicine, today.

[Vidya Venkatraman]

Hi David,

I am using TPP v5.1.0 and when i try to run xinteract on the Morpheus pep.xml, I get the following error that engine Morpheus not recognized. Any ideas if I am missing something?

xinteract (TPP v5.1.0 Syzygy, Build 201806211802-exported (Linux-x86_64))

running: "/hpc/apps/TPP/5.1.0/bin/InteractParser 'interact.morpheus.pep.xml' '2017-12-4_ABRF_200_DDA1.pep.xml' -L'7'"

 file 1: 2017-12-4_ABRF_200_DDA1.pep.xml

 processed altogether 72135 results

command completed in 9 sec 

running: "/hpc/apps/TPP/5.1.0/bin/DatabaseParser 'interact.morpheus.pep.xml'"

command completed in 1 sec 

running: "/hpc/apps/TPP/5.1.0/bin/RefreshParser 'interact.morpheus.pep.xml' '/common/venkatramanv/Data/Pratik/Thermo/DDA/raw/PRG2016_DECOY.fasta'"

  - Searching the tree...

  - Linking duplicate entries...

  - Printing results...

command completed in 15 sec 

running: "/hpc/apps/TPP/5.1.0/bin/PeptideProphetParser 'interact.morpheus.pep.xml' ACCMASS RT NONPARAM DECOYPROBS DECOY=DECOY_"

using Accurate Mass Bins

using RT

Using Decoy Label "DECOY_".

Decoy Probabilities will be reported.

Using non-parametric distributions

 (Morpheus)

error: engine Morpheus not recognized

command "/hpc/apps/TPP/5.1.0/bin/PeptideProphetParser 'interact.morpheus.pep.xml' ACCMASS RT NONPARAM DECOYPROBS DECOY=DECOY_" exited with non-zero exit code: 256

QUIT - the job is incomplete

Thanks

Vidya

[David Shteynberg]

Hello Vidya,

Currently, there is no support for Morpheus in the TPP.  If you are able to place you search results in a shared drive and send me a link for testing, I should be able to add the Morpheus support back in.

Thanks,

-Davod

To unsubscribe from this group and stop receiving emails from it, send an email to spctools-discuss+unsubscribe@googlegroups.com.
To post to this group, send email to spctools-discuss@googlegroups.com.

Visit this group at https://groups.google.com/group/spctools-discuss.
For more options, visit https://groups.google.com/d/optout.

[Vidya Venkatraman]

I just shared a box folder with you which contains mzML, fasta and the pepxml file.

https://cedars.app.box.com/folder/50657299852

Regards

Vidya

--
You received this message because you are subscribed to a topic in the Google Groups "spctools-discuss" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/spctools-discuss/qdmov2KGlJE/unsubscribe.
To unsubscribe from this group and all its topics, send an email to spctools-discuss+unsubscribe@googlegroups.com.

[David Shteynberg]

Hello Vidya,

I have added the support of Morpheus into the TPP codebase.  Please let me know which OS (windows or linux) you are processing on so I can provide you with an executable that you should be able to use.

Thanks,

-David

On Mon, Jun 25, 2018 at 1:09 PM, Vidya Venkatraman <vidya....@gmail.com> wrote:

I just shared a box folder with you which contains mzML, fasta and the pepxml file.

https://cedars.app.box.com/folder/50657299852

Regards

Vidya

[Vidya Venkatraman]

I am using Linux. Do i need to compile it again?

Regards

[David Shteynberg]

Which version of linux are you using?  I may be able to provide a binary for you.  Or you can recompile from source.

Cheers,

-David

On Tue, Jul 10, 2018 at 4:50 PM, Vidya Venkatraman <vidya....@gmail.com> wrote:

I am using Linux. Do i need to compile it again?

Regards

[Vidya Venkatraman]

I am using CentOS release 6.3

Regards

Vidya

[David Shteynberg]

Hello Vidya,

Unfortunately, I don't have that release easily available.  You will have to build the tools on your system.

Thanks,

-David

On Thu, Jul 12, 2018 at 10:14 PM, Vidya Venkatraman <vidya....@gmail.com> wrote:

I am using CentOS release 6.3

Regards

Vidya