Galaxy's Maxquant problem

igor · April 30, 2025, 10:52pm

Hi @Christos-Kotsopoulos

This part of the error message points on the issue:

It looks like data issue potentially related to the conversion:

github.com/cox-labs/CoxLab_Bug_Reporting

Different versions of MaxQuant analyze mzML files and Thermo RAW files

opened 11:50AM - 23 Apr 25 UTC

Yin-HJ

MaxQuant

**Describe the bug** I want to use MaxQuant to analyze mzML files converted from… Thermo RAW files in linux (I know MQ supports RAW files; in my case, I want to edit RAW files). I did some tests, here are the details 1. Initially, I used an earlier version of MQ (v.2.0.2.0) for analysis, and the raw format file analysis was smooth(because of the special database, I got about 1040 ms2 records in msms.txt). 2. However, when using converted mzML files (conversion made by msconvert in Windows and ThermoRawFileParser in Linux, the two converted mzML files are similar), MaxQuant reported an error at the Assemble step, and the error content is "**Could not Identify mass analyzer for spectrum 0._** _see Fig 1_". 3. Then, I tried to analyze using the latest version of MaxQuant (v2.6.7.0), and the process ended successfully, but I only got 11 ms2 records in msms.txt (mzML files converted from different methods got similar results). [ √ ] 4. After thatI tried to use MaxQuant (v2.6.7.0) to test the RAW format files, but an error occurred again. In the MSMS_preparation step, the error content was as follows: “**Arithmetic operation resulted in an overflow._ at MaxQuantLibS.Domains.Peptides.Data.PlistData.GenericPeakListLayerData.ReadCalibration()**... _see Fig 2_” ![Image](https://github.com/user-attachments/assets/dd814421-1066-47ef-9791-ccd78f4d30d5) **Attached files** For the configuration files, in addition to the basic experimental design, I checked GroupParameter-LFQ, Global Parameter-Label-free quantification - iBAQ, and Global Parameter - Identification - Match between runs, all using default parameters; The following are the mqpar.xmls for analysis [mqpar_v2.0.2.0.txt](https://github.com/user-attachments/files/19865801/mqpar_v2.0.2.0.txt) [mqpar_v2.6.7.0.txt](https://github.com/user-attachments/files/19865803/mqpar_v2.6.7.0.txt) Tested data are from [PRIDE-PXD006512](https://www.ebi.ac.uk/pride/archive/projects/PXD006512) - mzML converted by ThermoRawFileParse.exe in shell by: `mono ThermoRawFileParser.exe -i 20151115_liver_XL20_P_01.raw -b 20151115_liver_XL20_P_01.mzML -f 2` - mzML converted by msconvert in windows as follow Fig: ![Image](https://github.com/user-attachments/assets/c60033aa-b48d-4a3c-b7e6-dd1be7f46c4d) **Desktop** - OS: [Centos - 7.9.2009 ] - OS: [Windows11 - 23H2 ] **What I want to do** 1. I want to analyze the edited mass spectrum files in mzML format based on MaxQuant. 2. In addition, **why are the results of MaxQuant v2.6.7.0 analyzing mzML format files (about 1120 records in msms.txt) far different from analyzing RAW format files (about 11 records in msms.txt)?** **Additional context** 1. In fact, I suspected that there was a problem with the converted mzML format files; during the msconver conversion, I filtered the spectrum by "Vendor msLevel=1-2". The file size is close to that converted by ThermoRawFileParser.exe. Considering the successful execution of MaxQuant v2.6.7.0, it should not be the mzML file problem, I think。 2. By the way, if I do not use the "Vendor msLevel=1-2" filter, an error will be reported at the MSMS_preparation step analyzed by v2.0.2.0: ”Unable to read beyond the end of steam._ at System.IO.Stream.READAtLeastCore...“ (Yeah, error from assemble to MSMS_preparation). The same data analyzed by v2.6.7.0, pipeline ended successfully, I got the txt folder, but most files include msms.txt are empty, mzRange.txt contains many records. I am very confused and can provide the test RAW and mzML test data if needed. I'm not sure if there was a bug, or maybe I did something wrong that caused the error. Hope to get some help！ **I would be grateful for any suggestions or recommendations！** Yin HJ 2025/04/23 **Update 2025/04/24:** 1. I know why v2.6.7.0 could not analyze RAW files: because MaxQuant does not overwrite existing files, such as the index files, some files generated by v2.0.2.0 are used for analysis in V2.6.7.0. 2. Yestoday and today, I tried use MaxQuant v 2.6.7.0 with the same mzML format data, 1 success, 1 failure； On the other two computers, both v2.6.7.0 and v2.6.3.0 versions of MaxQuant failed, and the same error was reported for the three failures: "**Index was outside the bounds of the array._** " ![Image](https://github.com/user-attachments/assets/666bab1b-15bb-4f0d-ac23-0af75c34e390) 3. **There should be a bug for MaxQuant when analyzing mzML format files.**

Maybe try msconvert and other output formats, for example .mxXML.

Hope that helps.

Kind regards,
Igor