Skills matchms
🔬

matchms

Safe ⚡ Contains scripts⚙️ External commands🌐 Network access📁 Filesystem access

Analyze Mass Spectrometry Data for Metabolite ID

Also available from: davila7

Mass spectrometry data contains complex spectral information that requires specialized processing. Matchms provides comprehensive tools to import, filter, and compare mass spectra for metabolite identification and compound analysis.

Supports: Claude Codex Code(CC)
🥉 72 Bronze
1

Download the skill ZIP

2

Upload in Claude

Go to Settings → Capabilities → Skills → Upload skill

3

Toggle on and start using

Test it

Using "matchms". Calculate similarity between two spectra

Expected outcome:

  • Cosine similarity score: 0.85
  • Matching peaks: 45/89
  • Best matching m/z values: 147.076, 175.107, 203.138
  • Similarity assessment: High confidence match

Using "matchms". Load MGF file with 500 spectra

Expected outcome:

  • Loaded 500 spectra successfully
  • Average peaks per spectrum: 156
  • Precursor m/z range: 100.5 - 2000.3
  • Ion modes: positive (340), negative (160)

Using "matchms". Filter spectra with default filters

Expected outcome:

  • Applied 12 metadata harmonization filters
  • Normalized 500 intensity arrays
  • Removed 23 spectra with insufficient peaks
  • Final dataset: 477 spectra

Security Audit

Safe
v4 • 1/17/2026

All 268 static findings are false positives. The analyzer incorrectly flagged markdown code blocks (backticks) as shell execution, InChIKey descriptions as cryptographic algorithms, scientific database URLs as network reconnaissance, and legitimate Python code examples as malicious patterns. Matchms is a legitimate open-source mass spectrometry library for metabolomics research.

6
Files scanned
2,173
Lines analyzed
4
findings
4
Total audits

Risk Factors

⚡ Contains scripts (2)
references/filtering.md:237-273 references/workflows.md:75-502
⚙️ External commands (3)
SKILL.md:21-197 references/importing_exporting.md:11-392 references/similarity.md:7-367
🌐 Network access (3)
references/importing_exporting.md:407-416 references/similarity.md:380 references/workflows.md:645-647
📁 Filesystem access (1)
references/importing_exporting.md:323-344
Audited by: claude View Audit History →

Quality Score

45
Architecture
100
Maintainability
85
Content
29
Community
100
Security
91
Spec Compliance

What You Can Build

Identify unknown metabolites

Compare experimental spectra against reference libraries to identify unknown compounds.

Process LC-MS/MS data

Import raw mass spec data, apply quality filters, and prepare spectra for analysis.

Build spectral matching pipelines

Create automated workflows for large-scale spectral comparison and compound identification.

Try These Prompts

Basic spectrum loading 
Load spectra from my MGF file and show basic statistics about the dataset
Apply quality filters 
Apply default filters to normalize intensities and remove low-quality peaks from my spectra
Spectral library search 
Compare my query spectra against a reference library using cosine similarity and return top 5 matches
Build processing pipeline 
Build a SpectrumProcessor pipeline that applies metadata harmonization, intensity normalization, and peak filtering

Best Practices

  • Always apply default_filters first to harmonize metadata
  • Use appropriate similarity functions (CosineGreedy for speed, ModifiedCosine for precursor differences)
  • Validate results with chemical standards when possible

Avoid

  • Skipping quality filters before similarity calculations
  • Using inappropriate similarity metrics for your data type
  • Ignoring metadata standardization requirements

Frequently Asked Questions

What file formats does matchms support?
MGF, MSP, mzML, mzXML, JSON (GNPS), Pickle, and USI references.
Which similarity metric should I use?
Use CosineGreedy for general matching, ModifiedCosine when precursor masses differ.
Do I need RDKit for all features?
Only for chemical structure processing (SMILES, InChI conversions).
How do I handle large spectral libraries?
Use batch processing and consider memory-efficient similarity calculations.
Can matchms identify unknown compounds?
Yes, by comparing against reference spectral libraries using similarity scores.
What is the difference between mzML and MGF?
mzML contains raw instrument data, MGF is processed peak lists for analysis.

Developer Details

Author

K-Dense-AI

License

Apache-2.0 license

Repository

https://github.com/K-Dense-AI/claude-scientific-skills/tree/main/scientific-skills/matchms

Ref

main

File structure

📁 references/

📄 filtering.md

📄 importing_exporting.md

📄 similarity.md

📄 workflows.md

📄 SKILL.md