Trouble with passing results from searchgui to PeptideShaker

[justmpm]

Hi,
I have an issue with every .mzML or .mgf that are made with the latest version of Proteowizard (TPP compatibility, write index, filter top 100 peaks). They will work just fine with X!tandem in searchgui, but with SAGE or MS-GF+ they all fail. The search will complete but then PeptideShaker will not be able to open them. This is ture if searchgui automatically launches PeptideShaker or if I manually put in the completed search into PeptideShaker. I ran SAGE from the command line and PeptideShaker was also not able to open those results. The PeptideShaker log is pretty useless as it only gives the memory allocation at the beginning of the run. I will attach the html report that shows up in the output folder because it is more informative. At the moment, I will only put the reports from the SAGE search, because it is the one that I care the most about. The .txt contains a link to the .mzml.

The .mzML and (.mgf) work on other search engines just fine and I have taken the sage output and rescored it using MS2Rescore without any issue.

PeptideShaker Report 2026-02-03 17.42.01.html

SearchGUI Report 2026-02-03 17.42.11.html

mzml.txt

[hbarsnes]

That's strange. I just did a search in SearchGUI using Sage with your mzml file and I could open it in PeptideShaker without any issues. Would it be possible for you to also share the SearchGUI zip file you are trying to load? Plus the FASTA file?

[justmpm]

Thanks for checking for me. Here are the links to the files you asked for. The FASTA was downloaded from Uniprot and the decoys added by searchgui. Do you know what version of sage you are running? I tried to replace the the one in searchgui/resources/sage/windows with v0.14.7 but it didn't seem to make a difference. When it worked for you, did you see these java library warnings:

WARNING: A terminally deprecated method in sun.misc.Unsafe has been called
WARNING: sun.misc.Unsafe::objectFieldOffset has been called by io.airlift.compress.zstd.UnsafeUtil (file:/C:/PeptideShaker/PeptideShaker-3.0.13/lib/aircompressor-0.27.jar)
WARNING: Please consider reporting this to the maintainers of class io.airlift.compress.zstd.UnsafeUtil
WARNING: sun.misc.Unsafe::objectFieldOffset will be removed in a future release
WARNING: A restricted method in java.lang.System has been called
WARNING: java.lang.System::load has been called by org.sqlite.SQLiteJDBCLoader in an unnamed module (file:/C:/PeptideShaker/PeptideShaker-3.0.13/lib/sqlite-jdbc-3.45.2.0.jar)
WARNING: Use --enable-native-access=ALL-UNNAMED to avoid a warning for callers in this module
WARNING: Restricted methods will be blocked in a future release unless native access is enabled

I don't think they will break anything. I am running Temurin-25.0.1+8 (build 25.0.1+8-LTS)

Thanks again!
Mike

Fasta:
https://icgeborg-my.sharepoint.com/:u:/g/personal/myers_icgeb_org/IQA_wzKXAkBMSoNzp6rNZE9mAa2OkLhwqCt7ONQP7pKoHe8?e=NGZ5H7

searchgui result:
https://icgeborg-my.sharepoint.com/:u:/g/personal/myers_icgeb_org/IQBccUcrtuq-TKpdr7CBePHPAcjqXFQSDM9GDCIJztnj9zY?e=DcA7hG

[hbarsnes]

Thanks for sharing the files. I can load these without any issues. I'm using an older Java version though (21.0.8) and do not see the same warnings as you do. Maybe you can try with an older Java version to see of that solves the problem?

With regards to Sage, I'm just using the latest version added to SearchGUI, which seems to be version 0.14.6.

[justmpm]

Downgrading got rid of the java warnings. But things still aren't working for me. Same behavior with PeptideShaker quiting. I get the same behavior on another computer and I can't think of a reason why the mzml is working for you. I downloaded an mzml from a Q exactive and that is working just fine on my system. So it is pretty clear that it is something specific to my set up.

Thanks a lot for all your help!

Cheers
Mike

[justmpm]

Hi,
Can you tell me which versions of SearchGUI, PeptideShaker, and operating system you are using. I think that there is a subtle thing with the parsing of the spectrum files going on and maybe in a slightly different scenerio, there is not an issue with the Agilent data.

I remembered that X!Tandem doesn't really search .mzML files, it converts them into .mgfs and these might even be converted into an even simpler peaklists inside X!tandem. In Searchgui when you select X!Tandem as the search engine, the .mzML is converted into a .mgf that is briefly stored in searchgui/resources/temp/peaklists. If I take this temporary .mgf and feed it back into Searchgui, then it is able to complete a search using SAGE and successfully pass it to PeptideShaker with out any issues. The "same" data converted to a .mgf with either the Agilent software or msconvert will not successfully be analyzed by Sage when using SearchGUI as the wrapper for Sage. Let me know if you want me to upload the .mgfs, also please feel free to tell me if you don't have time to help me, since this seems to be an issue only I am having and that you cannot replicate on your end.

Cheers
Mike

[hbarsnes]

Hi Mike.

I'm using the latest releases of both tools, so I don't think there should be any differences there.

You are correct in your description of how mzML files are handled in SearchGUI. The reason being that not all search engines support mzML directly, hence we have to convert to mgf before giving the spectrum file to some of the search engines. Other search engines, including Sage, support mzML as input, thus removing the need for conversion. In such cases we are dependent on the search engine's internal mzML parsing which may also influence how it refers to the spectra in its output.

Have you compared the content of the two mgf files (i.e. internal SearchGUI mgf file vs. Agilent/msconvert)? Are there any obvious differences?

Best regards,
Harald

[justmpm]

Hi Harald,
The main difference is in the Title field. The file made by searchgui looks like:
BEGIN IONS
TITLE=scanId=2713
RTINSECONDS=2.70499999998
PEPMASS=352.193985711961 21910.631721496582
CHARGE=2+
84.07970703173942 239.50209045410156
85.02782350325617 268.0553894042969

The one from msconvert with the basic tpp compatible settings looks like:
BEGIN IONS
TITLE=lysateMEOH11.2713.2713.2 File:"lysateMEOH11.d", NativeID:"scanId=2713"
RTINSECONDS=2.705
PEPMASS=352.193985711961 21910.631721496582

CHARGE=2+
84.07970703 239.5020904541
85.0278235 268.0553894043

The "return" between pepmass and charge is in all the records from msconvert.

I can get a working mgf in msconvert by changing titleMaker. The cmd line looks something like: ProteoWizard 3.0.26021.4693c2b 64-bit\msconvert.exe" --mgf --noindex --filter "peakPicking vendor msLevel=1-" --filter "titleMaker "
The title looks like:
BEGIN IONS
TITLE=3
RTINSECONDS=2.705
PEPMASS=352.193985711961 21910.631721496582

CHARGE=2+
84.07970703 239.5020904541
85.0278235 268.0553894043

If I put anything other than Index into the titleMaker filter, PeptideShaker will throw an error that looks like:
Spectrum with title '"lysateMEOH11.3228401.3228401.3 File:""lysateMEOH11.d"", NativeID:""scanId=3228401"""' in file named 'lysateMEOH11' required to parse 'lysateMEOH11.sage.tsv.gz' not found.
Even replacing Index with Id will cause the error.

I can find that scan ID (3228401) in the mgfs made by msconvert, but not in the one created when Searchgui runs X!tandem. If I delete that spectrum from the mgf and rerun it, I get the same error, except it points at a different scan.

No matter what I do, the mzMLs made by msconvert will not work. Do you have any idea why the mzML will run on your machine, but not on mine. It looks like some sort of parsing error with PeptideShaker and searchgui having different tolerances, but I would guess that this error would also occur on your machine.

Cheers
Mike

here is a link to the mgf with full tpp compatibility settings : https://icgeborg-my.sharepoint.com/:u:/g/personal/myers_icgeb_org/IQDDv0FeGFhqT7EQm7qOAt5iAfwB1Mnsjm_3WykVj476J-8?e=Uq4bY1