search

Nesvilab / MSFragger

Ultrafast, comprehensive peptide identification for mass spectrometry–based proteomics
https://msfragger.nesvilab.org
103 stars 7 forks source link

Custom ion series #151

Closed telekandris closed 3 years ago

telekandris commented 3 years ago

I am trying to run a search with FragPipe v15.0 including a custom ion series (defined as "y C -249.0862"). When I include y in fragment_ion_series, it gives the error "ion type must be one of [a, b, c, b-18, b~, x, y, z, y-18, y~, Y]". When I do not include y (only in the definition field) the search runs until ProteinProphet is initiated, but it crashes because it does not find any PeptideProphet results. This error seems to be specific for custom ion series, because it does not happen if I do not define y at all. Could you suggest some workaround?

Thanks in advance!

fcyu commented 3 years ago

Are you using the latest MSFragger? I cannot reproduce your issue.

Best,

Fengchao

telekandris commented 3 years ago

I am. (MSFragger 3.2)

fcyu commented 3 years ago

This is weird. I put my log in the attachment. As you can see, MSFragger works well with your y* C -249.0862. Can you double check your settings?

Best,

Fengchao

log_2021-04-20_09-00-00.zip

telekandris commented 3 years ago

You just run MSFragger here, right? This also runs fur me. The error comes afterwards when initiating Protein Prophet.

dpolasky commented 3 years ago

Can you share your log file? It could be that the reason there aren't any results is that PeptideProphet modeling is failing

telekandris commented 3 years ago

Unfortunately it did not write the log into a file, but I checked and you are right, the model quality test fails for all (1+ to 7+)... What can I do about that?

danielgeiszler commented 3 years ago

Can you paste your PeptideProphet and ProteinProphet logs here so we can take a look?

On Apr 20, 2021, at 10:08 AM, telekandris @.***> wrote:

 Unfortunately it did not write the log into a file, but I checked and you are right, the model quality test fails for all (1+ to 7+)... What can I do about that?

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub, or unsubscribe.

telekandris commented 3 years ago

PeptideProphet [Work dir: C:\Users\Teresa\Desktop\currentanalysis\AT19\shiftion] C:\Users\Teresa\Downloads\FragPipe-14.0\fragpipe\lib..\tools\philosopher\philosopher.exe peptideprophet --nonparam --expectscore --decoyprobs --masswidth 1000.0 --clevel -2 --decoy rev --database C:\Users\Teresa\Downloads\FragPipe-14.0\fragpipe\tools\philosopher\2020-09-04-decoys-reviewed-contam-UP000005640.fas --combine 20210326_AT19_Rab_top.pepXML INFO[15:44:06] Executing PeptideProphet v3.4.13
Unknown file type. No file loaded.C:\Users\Teresa\Desktop\current_analysis\AT19\shiftion/20210326_AT19_Rab_top.raw Unknown file type. No file loaded.C:\Users\Teresa\Desktop\current_analysis\AT19\shiftion/20210326_AT19_Rab_top.raw file 1: C:\Users\Teresa\Desktop\current_analysis\AT19\shiftion\20210326_AT19_Rab_top.pepXML processed altogether 75360 results INFO: Results written to file: C:\Users\Teresa\Desktop\current_analysis\AT19\shiftion\interact.pep.xml

Using Decoy Label "rev_". Decoy Probabilities will be reported. Using non-parametric distributions (X! Tandem) (using Tandem's expectation score for modeling) using search_offsets in mass mixture distr: 0/329.0525 init with X! Tandem stricttrypsin PeptideProphet (TPP v5.2.1-dev Flammagenitus, Build 201906281613-exported (Windows_NT-x86_64)) AKeller@ISB read in 0 1+, 31929 2+, 37630 3+, 4330 4+, 1002 5+, 469 6+, and 0 7+ spectra. Found 36543 Decoys, and 38817 Non-Decoys MS Instrument info: Manufacturer: UNKNOWN, Model: UNKNOWN, Ionization: UNKNOWN, Analyzer: UNKNOWN, Detector: UNKNOWN

INFO: Processing standard MixtureModel ... Initialising statistical models ... Iterations: .........10.........20....WARNING: Mixture model quality test failed for charge (1+). . WARNING: Mixture model quality test failed for charge (2+). WARNING: Mixture model quality test failed for charge (3+). WARNING: Mixture model quality test failed for charge (4+). WARNING: Mixture model quality test failed for charge (5+). WARNING: Mixture model quality test failed for charge (6+). WARNING: Mixture model quality test failed for charge (7+). model complete after 26 iterations INFO[15:46:09] Done
Process 'PeptideProphet' finished, exit code: 0 Rewrite pepxml [Work dir: C:\Users\Teresa\Desktop\current_analysis\AT19\shiftion] java -cp C:\Users\Teresa\Downloads\FragPipe-15.0\fragpipe\lib/* com.dmtavt.fragpipe.util.RewritePepxml C:\Users\Teresa\Desktop\current_analysis\AT19\shiftion\interact.pep.xml C:\Users\Teresa\Desktop\current_analysis\AT19\20210326_AT19_Rab_top.raw Fixing pepxml: C:\Users\Teresa\Desktop\current_analysis\AT19\shiftion\interact.pep.xml Writing output to: C:\Users\Teresa\Desktop\current_analysis\AT19\shiftion\interact.pep.xml5348710383317590743.temp-rewrite Deleting file: C:\Users\Teresa\Desktop\current_analysis\AT19\shiftion\interact.pep.xml Moving rewritten file to original location: [C:\Users\Teresa\Desktop\current_analysis\AT19\shiftion\interact.pep.xml5348710383317590743.temp-rewrite] -> [C:\Users\Teresa\Desktop\current_analysis\AT19\shiftion\interact.pep.xml] Process 'Rewrite pepxml' finished, exit code: 0 ProteinProphet [Work dir: C:\Users\Teresa\Desktop\current_analysis\AT19\shiftion] C:\Users\Teresa\Downloads\FragPipe-14.0\fragpipe\lib..\tools\philosopher\philosopher.exe proteinprophet --maxppmdiff 2000000 --output combined C:\Users\Teresa\Desktop\current_analysis\AT19\shiftion\interact.pep.xml INFO[15:46:11] Executing ProteinProphet v3.4.13
ProteinProphet (C++) by Insilicos LLC and LabKey Software, after the original Perl by A. Keller (TPP v5.2.1-dev Flammagenitus, Build 201906281613-exported (Windows_NT-x86_64)) (no FPKM) (using degen pep info) Reading in C:/Users/Teresa/Desktop/current_analysis/AT19/shiftion/interact.pep.xml... did not find any PeptideProphet results in input data! Did you forget to run PeptideProphet? ...read in 0 1+, 0 2+, 0 3+, 0 4+, 0 5+, 0 6+, 0 7+ spectra with min prob 0.05

WARNING: no data - output file will be empty FATA[15:46:13] Cannot execute program. There was an error with ProteinProphet, please check your parameters and input files Process 'ProteinProphet' finished, exit code: 1 Process returned non-zero exit code, stopping

dpolasky commented 3 years ago

It looks like PeptideProphet is not able to distinguish targets from decoys and can't make sense of the search results, so it does not output anything. I suspect this is happening only when using the custom y ion because it introduces too much noise into the search. Given that you are doing a mass offset search, are you trying to find a labile modification that might leave behind this remainder mass on the y ions? If so, I think what's happening is the custom y ion is searched for ALL spectra, not just those with the mass offset of the modification, and this is causing trouble in matching the rest of the unmodified spectra. Are you still including regular b/y ions in the search? If not, that would definitely cause this problem. If so, it will likely be challenging to get this to converge with the custom y ions - if that's the case, maybe you can contact us directly via email to discuss settings for a labile PTM search like this in more detail

telekandris commented 3 years ago

This is exactly the case, I suspected that searching every spectra for the shifted ions might cause problems, but I could not find a simple way to restrict the search. Thanks for explaining, I will come back to you after discussing it with my boss.