PTMProphet MSFragger modification mass discrepancy

48 views
Skip to first unread message

Alex Zelter

unread,
Mar 18, 2020, 2:38:51 PM3/18/20
to spctools-discuss
I am running PTMProphet (TPP v5.2.1-dev Flammagenitus, Build 202003100907-8034) on MSFragger (MSFragger version MSFragger-2.3) output and ending up with strange results in specific cases.

For example, for a specific PSM, 52782, MSFragger outputs:

<spectrum_query start_scan="52782" uncalibrated_precursor_neutral_mass="2288.8486" assumed_charge="3" spectrum="QEP2_2018_0812_AZ_033_az736_AZ.52782.52782.3" end_scan="52782" index="41840" precursor_neutral_mass="2288.8384" retention_time_sec="6104.521">
<search_result>
<search_hit peptide="RYKAAFTECCQAADK" massdiff="471.0195" calc_neutral_pep_mass="1817.8188" peptide_next_aa="A" num_missed_cleavages="2" num_tol_term="2" num_tot_proteins="5" tot_num_ions="56" hit_rank="1" num_matched_ions="23" protein="sp|P02768|ALBU_HUMAN Serum albumin OS=Homo sapiens GN=ALB PE=1 SV=2" peptide_prev_aa="K" is_rejected="0">
<modification_info>
<mod_aminoacid_mass mass="160.0307" position="9"/>
<mod_aminoacid_mass mass="160.0307" position="10"/>
</modification_info>
<search_score name="hyperscore" value="44.421"/>
<search_score name="nextscore" value="12.357"/>
<search_score name="expect" value="4.415e-15"/>
</search_hit>
</search_result>
</spectrum_query>

This would indicate that C9 is +57, C10 is +57 and there is an additional unlocalized mass diff of +471.

After running TPP I end up with output like this in interact.ipro.pep.xml for that scan:

<spectrum_query start_scan="52782" uncalibrated_precursor_neutral_mass="2288.8486" assumed_charge="3" spectrum="QEP2_2018_0812_AZ_033_az736_AZ.52782.52782.3" end_scan="52782" index="36796" precursor_neutral_mass="2288.8384" retention_time_sec="6104.521">
<search_result>
<search_hit peptide="RYKAAFTECCQAADK" massdiff="471.0195" calc_neutral_pep_mass="1817.8188" peptide_next_aa="A" num_missed_cleavages="1" num_tol_term="2" num_tot_proteins="5" tot_num_ions="56" hit_rank="1" num_matched_ions="23" protein="sp|P02768|ALBU_HUMAN" peptide_prev_aa="K" is_rejected="0" protein_descr="Serum albumin OS=Homo sapiens GN=ALB PE=1 SV=2">
<modification_info modified_peptide="RYKAAF[618]TEC[160]C[160]QAADK">
<mod_aminoacid_mass position="6" mass="618.0879"/>
<mod_aminoacid_mass position="9" mass="574.0287"/>
<mod_aminoacid_mass position="10" mass="574.0287"/>
</modification_info>.......

So it seems to me that we now have a modification mass addition of +471 on residues 6 9 and 10 rather than a mass addition of +471 on residue 6 and then an addition of +57 on residues 9 and 10.

In cases where there is just a mass diff (open modification) and no defined variable modifications output is as expected. This situation seems to occur when both defined variable modifications and open modifications are present.

I have confirmed the same behavior in older versions of MSFragger and TPP.

Any ideas would be much appreciated!
Thanks,
Alex

David Shteynberg

unread,
Mar 18, 2020, 2:42:22 PM3/18/20
to spctools-discuss
Hi Alex,

This seems like a bug that should be fixed.  Are you running PTMProphet at any point in your analysis pipeline?

Thanks,
-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 view this discussion on the web visit https://groups.google.com/d/msgid/spctools-discuss/5a37d860-cd21-445a-a485-ed3684291f4d%40googlegroups.com.

Alex Zelter

unread,
Mar 18, 2020, 2:56:17 PM3/18/20
to spctools-discuss
Hi David,
Thanks for the fast response!

My list of commands are as follows (after running MSFragger):

tpp/bin/InteractParser       interact.pep.xml FraggerOutput.pepXML
tpp/bin/DatabaseParser       interact.pep.xml
tpp/bin/RefreshParser        interact.pep.xml /path/to/database/fastaFile.fasta
tpp/bin/PeptideProphetParser interact.pep.xml ACCMASS DECOYPROBS DECOY=random NONPARAM MASSWIDTH=520 MINPROB=0
tpp/bin/InterProphetParser   interact.pep.xml interact.ipro.pep.xml
tpp/bin/PTMProphetParser MASSDIFFMODE interact.pep.xml
tpp/bin/PTMProphetParser MASSDIFFMODE interact.ipro.pep.xml

I'm happy to share the full set of files with you if that helps.

Thanks!
Alex
To unsubscribe from this group and stop receiving emails from it, send an email to spctools...@googlegroups.com.

David Shteynberg

unread,
Mar 18, 2020, 3:00:34 PM3/18/20
to spctools-discuss
Thanks Alex.  I suspect it is a possible bug in PTMProphet. Can you see what step of your analysis created these additional mods on C?  I would help me narrow down the places to look.

Cheers,
David

To unsubscribe from this group and stop receiving emails from it, send an email to spctools-discu...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/spctools-discuss/4314cc4f-3944-4a51-bb34-dc43aae61848%40googlegroups.com.

Alex Zelter

unread,
Mar 24, 2020, 6:59:55 PM3/24/20
to spctools-discuss
Hi David,
Your v5.2.1-dev Flammagenitus, Build 202003241419-80410 has fixed this problem.
Thank you!
Alex
Reply all
Reply to author
Forward
0 new messages