Questions regarding PTM Prophet parameters

[Vidya Venkatraman]

Hi David,

I am running the TPP v4.6 OCCUPY rev 3, Build 201307241109 (MinGW) version of TPP. I have recently started exploring PTM prophet in TPP. I had a few questions regarding the options and parameters within PTM Prophet. 

Feel free to point me to any documentation or tutorial if there is one already on these parameters. I can also send any input or result files if needed. 

1) The default residue mass shift of 79.96 on STY does not seem to work in my case. Although, if i give the total mass change of 167 for S, 181 for T and 243 for Y, it works. I noticed that my pep.xml files mention total masses. Should total mass change be used?

This is the error I get:

running: "C:/Inetpub/tpp-bin/PTMProphetParser STY,79.966 MZTOL=0.1 interact.ipro.pep.xml interact.ptm.ipro.pep.xml"
INFO: Writing file interact.ptm.ipro.pep.xml ...
INFO: Reading file interact.ipro.pep.xml ...
	WARNING: Unrecognized mod on peptide: MAQPAWGLSASHRAPAPPPPPLLLPLPLLLPLLELPLLHRS[167]S

I also noticed that even if i use total mass change, for one peptide, I got this warning

running: "C:/Inetpub/tpp-bin/PTMProphetParser S,167,T,181,Y,243 MZTOL=0.1 interact.ipro.pep.xml interact.ptm.ipro.pep.xml"
INFO: Writing file interact.ptm.ipro.pep.xml ...
INFO: Reading file interact.ipro.pep.xml ...
	WARNING: Unrecognized mod on peptide: IPESLGLMTLPPESLQKT[181]
command completed in 1 sec 

2) Should the m/z tolerance in the PTM prophet parameters be the same as the fragment tolerance used during the search? If not, What does it exactly represent?

3) What is the "Do not update modification_info tags in pepXML" option meant for?

4) Within the interact.ptm.ipro.pep.xml, what does the tag "ptmprophet_result prior="0.5" mean?

<analysis_result analysis="ptmprophet">
<ptmprophet_result prior="0.5" ptm="PTMProphet_T79.9663" ptm_peptide="IPESLGLMT(0.916)LPPESLQKT(0.084)">
<mod_aminoacid_probability position="9" probability="0.916"/>
<mod_aminoacid_probability position="18" probability="0.084"/>
</ptmprophet_result>

5) The current TPP workflow seems to have options to run ProteinProphet after PeptideProphet and/or after InterProphet. I was wondering if it was possible to run ProteinProphet after PTMProphet? (i.e. can we run ProteinProphet on the interact.ptm.ipro.pep.xml file created by PTMProphet). Are there any reasons as to why that should not be done or cannot be done after PTMProphet?

I would highly appreciate any inputs or suggestions from your end.

Thanks in advance,
Vidya

[David Shteynberg]

The proper way to run PTMProphet is with the mass shift and grouping amino acids sharing that shift.  STY,79.966 is correct.  Are you using any other mods in your search...variable or static?  A mass mod that is unknown to PTMProphet is causing the warning, can you look for this peptide ti the input files to see if it has any other mods on it?

Tolerance should be same as search fragment tolerance.

Do not update will keep the orignal modifications call made by search algorithm instead of updating to PTMProphet predicted sites.

Prior mean prior probabilty that each site is modified without looking at peaks but assuming the IDis correct.  If a peptide has 3 sites and one mod prior is 1/3.

ProteinProphet can be run on PTMProphet containing pepXML file.  The peptides should display the PTMProphet mod predictions unless the no upgrade option youasked about is enabled.

Hope this helps,

David

--
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.

[Vidya Venkatraman]

Hi David,

Thanks for your prompt response. I did have variable mod on Oxidation and fixed mod on Carbamidomethyl. Adding these along with the STY, 79.966 seems to have fixed the problem I was having. 

I noticed one thing that if I select the option "Run ProteinProphet afterwards" after PeptideProphet and If i also select the option "Also run ProteinProphet on these results" after InterProphet, I seem to get "c:/Inetpub/wwwroot/ISB/data/ptmtest/interact.ipro.prot.xml (file not found: not created or deleted?)" in the output. (please find attached screenshot of output and logs). Should this be expected? Should this be run differently?

Thanks
Vidya