MaxQuant: Install, Run and Read proteinGroups.txt

MaxQuant is a proteomics software package for analysing large mass-spectrometric data sets, developed in the Cox lab and freely downloadable. This guide is for researchers who have raw LC-MS/MS files and a protein FASTA database and need to get from download to a clean protein table: what the program needs, the run steps, which output file to open first, and which columns to filter before any statistics.

What MaxQuant is used for in research

The MaxQuant documentation describes it as a proteomics software package designed for analysing large mass-spectrometric data sets. It supports labelled and label-free quantification, includes the Andromeda search engine and a viewer, and hands statistical analysis to the Perseus framework. Background on the underlying method is in what proteomics is, MS/MS spectrometry and protein identification by database search.

Install or access (official source only)

Download MaxQuant.zip from the official site (maxquant.org, with documentation at cox-labs.github.io/coxdocs) and unpack it. The documentation states that downloading and using the software is free of charge.

Item	Requirement stated in the Cox lab installation page
Memory	4 GB RAM minimum, and 4 GB RAM per thread executed in parallel
Windows	Windows 10 or 11, or Windows Server 2016, 2019 or 2022 (64-bit)
.NET	.NET Core 8.0 or higher; at least 8.0.302 is recommended because of GUI bug fixes; .NET 3.1 and older are no longer supported
Linux	Ubuntu 20.04 or higher; command line only, no GUI support at present

On Linux, check the runtime with dotnet --version, then run dotnet MaxQuant/bin/MaxQuantCmd.dll mqpar.xml, where mqpar.xml is the parameter file.

Step-by-step: a first search in the Windows GUI

The official First Steps walkthrough was written for an older MaxQuant release (about 1.4.3.14, 2014), so tab layouts in the current GUI can differ slightly. The sequence below follows that walkthrough; most default parameters are fine for a first run.

1. Unpack MaxQuant.zip and double-click MaxQuant.exe.
2. On the Raw files tab, use Load for individual files or Load folder for a directory of raw files.
3. Specify the experimental design for the loaded files.
4. In the group-specific parameters, keep label type Standard and set Multiplicity: 1 for label-free, 2 for light/heavy, 3 for light/medium/heavy. Move the enzyme you used (the walkthrough uses Trypsin/P) to the selected list.
5. In the global parameters, use Add file to select the FASTA database.
6. Click Start (lower left). The Performance tab shows progress, and a popup reports Done.

Output and how to read it

Results are written as tab-delimited text in the combined\txt folder (the combined folder, then txt inside it) under the raw-file location; combined\txt\tables.pdf describes every column. According to the output-tables documentation, evidence.txt is usually the only file needed for downstream work, and the other main tables are peptides.txt, proteinGroups.txt, modificationSpecificPeptides.txt, msms.txt and aifMsms.txt, plus a summary table.

Filter these columns first

Column	What the documentation says	Action
Reverse	A + marks a decoy hit	Remove rows marked + before further analysis
Potential contaminant	A + marks a likely contaminant	Remove these rows as well
Only identified by site	Present in the protein groups table	Check it alongside the two columns above
LFQ intensity [experiment]	Per-experiment label-free quantity column	Present in proteinGroups.txt
iBAQ [experiment]	Per-experiment column	Present in proteinGroups.txt

Statistical analysis is normally continued in Perseus, the framework the MaxQuant documentation names for that purpose.

Troubleshooting

• The program will not start or the GUI misbehaves: confirm a 64-bit OS and .NET 8.0 or higher; the documentation recommends at least 8.0.302 for GUI fixes, and .NET 3.1 or older is unsupported.
• The run is slow or fails on large data: the documentation requires 4 GB RAM per parallel thread, so reduce the number of threads if memory is short.
• No GUI on Linux: this is expected; run MaxQuantCmd.dll with a parameter file as shown above.
• Which file do I open? Start with proteinGroups.txt for protein-level tables, or evidence.txt, which the documentation calls usually the only file needed.

Free alternatives

Tool	What the CASRAI guide reports
FragPipe	Free-to-download suite from the Nesvilab with a GUI and command line on Windows and Linux; its search engine is MSFragger. Academic and commercial licence terms are on separate MSFragger wiki pages.

For comparison, Spectronaut is commercial DIA software, and the guide notes it can import external search results from MaxQuant. The historical search algorithm is covered in SEQUEST.

FAQ

Is MaxQuant free?

The official installation page states that downloading and using the software is free of charge.

Does MaxQuant run on Linux?

Yes, on Ubuntu 20.04 or higher, but only from the command line with .NET installed; the GUI is not supported on Linux at present.

How much memory does MaxQuant need?

At least 4 GB of RAM, plus 4 GB per thread run in parallel.

Which MaxQuant output file should I use?

The documentation says evidence.txt is usually the only file needed; proteinGroups.txt holds the protein-level table. Remove Reverse and Potential contaminant rows first.

Sources (official docs)

• Cox lab, MaxQuant documentation: Download and installation, First steps, Output tables, MaxQuant instructions (read 2026-11-08).
• maxquant.org (read 2026-11-08).