How to Use MaxQuant — A Step-by-Step Tutorial for Beginners

Introduction

MaxQuant is the most widely used free software for processing mass spectrometry-based proteomics data. Developed by Jürgen Cox's group at the Max Planck Institute, it has been cited in over 10,000 publications and remains the go-to tool for many proteomics researchers.

Despite its power, MaxQuant can be intimidating for beginners. This step-by-step tutorial will guide you through installing, configuring, and running MaxQuant for a typical label-free DDA proteomics experiment.

Prerequisites

Before starting, you'll need:

• Windows PC (MaxQuant runs natively on Windows; Linux via Mono is possible but less stable)
• 8+ GB RAM (16+ GB recommended for large datasets)
• Raw files from a Thermo, Bruker, or other supported instrument
• FASTA database of your organism's proteome from UniProt

Step 1: Download and Install MaxQuant

1. Go to maxquant.org
2. Register for a free account
3. Download the latest version (2.6.x as of 2026)
4. Extract the ZIP file to a folder (e.g., C:\MaxQuant)
5. Run MaxQuant.exe — no installation needed

Tip: Keep MaxQuant on a fast drive (SSD). The software writes temporary files that benefit from fast I/O.

Step 2: Download Your FASTA Database

1. Go to UniProt
2. Click Proteomes → Search for your organism (e.g., "Homo sapiens")
3. Select the reference proteome (UP000005640 for human)
4. Download as FASTA (canonical) — this gives you one sequence per gene (~20,400 entries for human)
5. Save to a dedicated folder (e.g., C:\Databases\)

Important: Include the MaxQuant contaminants database (automatically added by MaxQuant) and do NOT add decoy sequences manually — MaxQuant generates these internally.

Step 3: Load Your Raw Files

1. In MaxQuant, click the Raw files tab
2. Click Load and select your .raw files (or drag and drop)
3. Set the Experiment column to group biological replicates:
   • e.g., "Control" for control samples, "Treatment" for treated samples
4. Set Fractions if you ran fractionated samples (otherwise leave as default)

File Organization Tips

• Keep raw files on a local drive (not network storage) for best performance
• Use consistent, descriptive file names
• Create a separate folder for each project

Step 4: Configure Group-Specific Parameters

Click the Group-specific parameters tab. These settings apply to each experimental group.

Type

• Standard for label-free experiments
• SILAC if you used metabolic labeling
• Select TMT labels if using TMT/iTRAQ

Digestion

• Enzyme: Trypsin/P (default, correct for most experiments)
• Max missed cleavages: 2 (standard)

Modifications

Fixed modifications:

• Carbamidomethyl (C): Add this if you used iodoacetamide for cysteine alkylation (standard in most protocols)

Variable modifications:

• Oxidation (M): Always include (methionine oxidation occurs during sample prep)
• Acetyl (Protein N-term): Recommended to include
• Add others only if specifically relevant to your experiment (e.g., Phospho (STY) for phosphoproteomics)

Tip: Each variable modification multiplies the search space. Keep them to a minimum (3-4 max) for standard experiments.

Instrument

• Select your instrument type from the dropdown
• Mass tolerance is set automatically but can be adjusted:
  • First search: 20 ppm (for mass recalibration)
  • Main search: 4.5 ppm (after recalibration — this is usually optimal)

Step 5: Configure Global Parameters

Click the Global parameters tab.

Sequences

1. Click Add under the FASTA files section
2. Navigate to your downloaded FASTA file and select it
3. MaxQuant will automatically add its built-in contaminants database

Identification

• PSM FDR: 0.01 (1% — standard)
• Protein FDR: 0.01 (1% — standard)
• Min peptide length: 7 (default)
• Min peptides for identification: 1 (use 2 for more stringent results)
• Match between runs: Enable this if you have multiple fractions or replicates (transfers identifications between runs to reduce missing values)

Quantification

For label-free quantification:

• LFQ: Check "Label-free quantification"
• LFQ min ratio count: 2 (default)
• iBAQ: Check if you want intensity-based absolute quantification (useful for estimating protein copy numbers)

Advanced Settings

• Number of threads: Set to the number of CPU cores available (e.g., 8 for an 8-core processor)
• Keep temporary files: Uncheck to save disk space (can delete 50+ GB of temp files)

Step 6: Run the Analysis

1. Click the Start button (green play icon)
2. MaxQuant will process your data in several stages:
   • Feature detection: Identifying peptide peaks in MS1
   • First search: Rough identification for mass recalibration
   • Main search: Full database search using Andromeda
   • FDR calculation: Using target-decoy approach
   • Quantification: LFQ normalization and ratio calculation
   • Match between runs: If enabled

Expected Processing Time

Dataset Size	Processing Time
10 raw files	2-6 hours
50 raw files	12-24 hours
100+ raw files	1-3 days

Tips for faster processing:

• Use an SSD for raw files and temp directory
• Maximize available RAM
• Close other applications
• Consider running overnight for large datasets

Step 7: Understanding the Output

MaxQuant generates output in the combined/txt/ folder. Key files:

proteinGroups.txt

The main results file containing one row per protein group:

• Protein IDs: UniProt accessions
• Gene names: Human-readable gene symbols
• LFQ intensity [experiment]: Normalized protein quantities
• iBAQ: Absolute abundance estimates
• Sequence coverage: Percentage of protein covered by identified peptides
• Razor + unique peptides: Number of peptides assigned to each protein
• Reverse: "+" means this is a decoy hit (filter out)
• Potential contaminant: "+" means this is a common contaminant (filter out)
• Only identified by site: "+" means identified only by a modification (usually filter out)

peptides.txt

Peptide-level results:

• Sequences, modifications, intensities, scores, and mapping to proteins

evidence.txt

Most detailed output — one row per peptide feature per raw file:

• Retention times, m/z values, intensities, charge states

msms.txt

MS2 spectrum information:

• Fragmentation data, scores, identifications

summary.txt

Overview statistics per raw file:

• Number of MS1 and MS2 scans, identified peptides and proteins

Step 8: Downstream Analysis with Perseus

Perseus is MaxQuant's companion tool for statistical analysis.

Loading Data

1. Download Perseus from maxquant.org/perseus
2. Load proteinGroups.txt: Generic matrix upload → select the file
3. Main columns: Select LFQ intensity columns
4. Categorical columns: Include Gene names, Protein IDs

Standard Filtering

Processing → Filter rows based on categorical column:
  - Remove "Reverse" = "+"
  - Remove "Only identified by site" = "+"
  - Remove "Potential contaminant" = "+"

Log2 Transformation

Processing → Basic → Transform → log2(x)

Filter for Valid Values

Processing → Filter rows → Filter rows based on valid values:
  - Min values: 3 (or 70% of samples per group)
  - Mode: "In at least one group"
  - Grouping: Define your experimental groups

Missing Value Imputation

Processing → Imputation → Replace missing values from normal distribution:
  - Width: 0.3
  - Down shift: 1.8

Statistical Testing

Processing → Tests → Two-sample test (t-test):
  - S0: 0.1 (or 1 for more conservative)
  - FDR: 0.05 (Permutation-based)
  - Number of randomizations: 250

Visualization

• Volcano plot: Shows fold change vs. significance
• Heatmap: Clustering → Hierarchical clustering
• PCA: Dimensionality reduction → Principal component analysis

Common Troubleshooting

"No identifications"

• Check your FASTA file matches your organism
• Verify the enzyme setting is correct
• Ensure fixed modifications match your sample preparation
• Check that raw files are not corrupted

Very few identifications

• Mass tolerance might be wrong — check the summary.txt for mass error distribution
• Sample quality may be low — check TIC in the viewer
• Database may be wrong species

MaxQuant crashes

• Increase available RAM
• Reduce number of threads
• Update to latest version
• Check for corrupted raw files

Match between runs issues

• Only works within the same experiment
• Can introduce false transfers — validate carefully
• Disable if comparing very different sample types

Best Practices Summary

1. Always use the latest MaxQuant version
2. Keep your FASTA database updated (download fresh from UniProt annually)
3. Use appropriate controls (blank runs, QC standards)
4. Filter properly in Perseus (reverse, contaminants, only-by-site)
5. Use at least 3 biological replicates per condition
6. Document your parameters — save the mqpar.xml file for reproducibility
7. Check QC metrics before proceeding to statistical analysis

Beyond MaxQuant: When to Consider Alternatives

While MaxQuant is excellent, consider alternatives for specific use cases:

• DIA data: Use DIA-NN or Spectronaut instead
• Very large datasets (>500 files): FragPipe/MSFragger is faster
• Open modification searches: MSFragger excels here
• TMT large-scale: FragPipe handles TMT more efficiently

Conclusion

MaxQuant remains one of the best free tools for proteomics data analysis. By following this tutorial, you should be able to go from raw mass spectrometry files to a list of quantified proteins ready for biological interpretation.

The learning curve is real, but the proteomics community is supportive. The MaxQuant Google Group, annual MaxQuant Summer School, and Nature Protocols publications are excellent resources for deepening your skills.

Start with a small dataset, understand each parameter, and build from there. Happy analyzing!

---

Related Reading

• 🔗 How to Analyze Mass Spectrometry Data
• 🔗 DIA-NN Proteomics Software Review
• 🔗 DIA vs DDA in Proteomics
• 💊 Exercise and Brain Health Neuroscience
• 💻 Best Budget Laptops 2026 Under 1000