MatchMS

What it does

MatchMS adds robust mass spectrometry data processing capabilities: import/export across common formats (mzML, MGF, MSP, JSON), normalize and filter spectra, harmonize metadata, derive chemical identifiers (InChI/SMILES), and compute spectral similarity scores (cosine, modified cosine, fingerprint-based). It is designed to build reproducible analysis pipelines for library matching, quality control, and large-scale similarity comparisons.

When to use it

Use this skill when you need to preprocess raw MS data, standardize metadata across files, run peak filtering and normalization, or perform spectral library matching and scoring. Typical scenarios include metabolomics identification, spectral networking, and automated library searches for unknown spectra.

What's included

• Scripts: reference examples for importing/exporting, filtering, processing pipelines (has_references=true).
• References: detailed docs for filtering, similarity metrics, and workflows under references/.
• Instructions: examples show how to load spectra, apply default_filters, normalize intensities, require minimum peaks, build a SpectrumProcessor pipeline, and calculate similarity scores with Cosine/ModifiedCosine.

Compatible agents

This skill is implementation-agnostic but is most useful to Python-enabled agents or Claude/Copilot-style agents that can produce/run Python code and work with scientific packages.