Similarity scoring for spectra

Skill HolobiomicsLab/asb-skill-collections/collections/metabolomics/v2/skills/similarity-scoring-for-spectra

Use when you have LC-MS/MS query spectra in mgf format that you need to match against a custom database (e.g., prepared with CFM-id) to identify compounds. Apply this skill when you want to rank candidate compounds by spectral similarity and return scored match results for downstream interpretation.From its SKILL.md

Install

npx -y skills add HolobiomicsLab/asb-skill-collections --skill similarity-scoring-for-spectra

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

15 stars15 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its file declares

Copied from the file, not written here

The file declares its own license as CC-BY-4.0. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.

SKILL.md

6.1 KB, 901 tokens by cl100k_base, as published. Nobody here has run it

similarity-scoring-for-spectra

Summary

Compute spectral similarity scores between query MS/MS spectra (in Mascot Generic Format) and a reference database to identify compounds. This skill quantifies the degree of spectral match to enable compound assignment in metabolomics workflows.

When to use

You have LC-MS/MS query spectra in mgf format that you need to match against a custom database (e.g., prepared with CFM-id) to identify compounds. Apply this skill when you want to rank candidate compounds by spectral similarity and return scored match results for downstream interpretation.

When NOT to use

Input spectra are not in Mascot Generic Format (mgf) — convert to mgf first.
Reference database was not prepared using CFM-id; MS2Compound v1.0.2 is optimized for CFM-id–derived databases.
You need to identify novel compounds with no reference database; this skill requires a pre-built custom database.

Inputs

Mascot Generic Format (mgf) query file containing LC-MS/MS spectra
Custom compound database prepared with CFM-id (fragment prediction tool)
MS2Compound v1.0.2 GUI configuration

Outputs

Compound identification results file with match scores
Ranked list of candidate compounds per query spectrum
Metadata and similarity metrics for matched compounds

How to apply

Load the custom compound database (prepared using CFM-id fragment predictions) into MS2Compound v1.0.2. Import the mgf query file containing LC-MS/MS spectra. Execute the matching algorithm, which compares each query spectrum against database entries and computes similarity scores for all candidate compounds. Export the identified compounds with their match scores and associated metadata. The algorithm's output ranks compounds by similarity strength, allowing prioritization of high-confidence identifications.

Related tools

MS2Compound v1.0.2 (Graphical user interface that implements the spectral matching algorithm and similarity score computation) — https://sourceforge.net/projects/ms2compound/
CFM-id (Fragment prediction tool used to prepare the custom reference database that MS2Compound searches against)

Evaluation signals

Match scores are computed for all query spectra against all database entries (no missing values in results file).
Compound rankings are consistent across repeated runs on identical inputs (deterministic algorithm).
Results file contains structured metadata including spectrum identifiers, candidate compound names, and similarity scores for each match.
High-confidence matches (top-ranked compounds with highest similarity scores) correspond to visually similar precursor m/z and fragment patterns in the input mgf.
Score distribution shows variation across candidates, confirming discrimination between similar and dissimilar compounds.

Limitations

MS2Compound is compatible only with custom databases prepared using CFM-id; incompatible with databases from other fragment prediction tools.
Tool requires Windows operating system (Windows 7, 8, or 10); Linux/macOS deployments are not documented in the README.
No changelog provided in README; algorithm updates and improvements are not explicitly documented.
Minimum system configuration (2 GB RAM, Intel i3 64-bit CPU) may limit performance on large databases or high-throughput batches.

Evidence

[readme] MS2Compound (v1.0.2) is a user friendly Graphical User Interface (GUI) for the identification of the compounds from LC-MS and MS/MS metabolomics data.: "MS2Compound (v1.0.2) is a user friendly Graphical User Interface (GUI) for the identification of the compounds from LC-MS and MS/MS metabolomics data."
[readme] The current version of MS2Compound is compatible with the customized database prepared using CFM-id, the fragment prediction tool. Mascot Generic Format (mgf) files can be used as query input file.: "The current version of MS2Compound is compatible with the customized database prepared using CFM-id, the fragment prediction tool. Mascot Generic Format (mgf) files can be used as query input file"
[other] Execute the matching algorithm to compare query spectra against database entries and compute similarity scores.: "Execute the matching algorithm to compare query spectra against database entries and compute similarity scores."
[other] Export identified compounds with their match scores and metadata to a results file.: "Export identified compounds with their match scores and metadata to a results file."

What ships with it

Read from the repository

Just SKILL.md. No reference files, no scripts.