matchms

You can identify unknown compounds from mass spectrometry data by importing your experimental spectra and comparing them against a known spectral library using similarity metrics like cosine or modified cosine algorithms.

You can import mass spectrometry data for spectral analysis from MGF, mzML, and MSP formats, allowing you to standardize and filter spectra from various experimental sources before processing.

Cosine similarity measures direct peak overlap between spectra, while modified cosine similarity accounts for precursor mass shifts, enabling accurate library matching even when analyzing derivatives or modified metabolites.

Yes, you can filter and standardize mass spectrometry spectra before library searching by processing imported data to remove noise, normalize peak intensities, and ensure consistent format across samples.

Fingerprint-based spectral comparison improves metabolite identification by analyzing structural fingerprints alongside mass peaks, providing an advanced similarity metric that complements traditional cosine scoring for complex compound matching.

The best way to compare an unknown spectrum against a spectral library is to import both datasets, apply similarity metrics such as cosine or modified cosine algorithms, and rank the top matching compounds by score.