peptide-evidence-map

Deterministic normalization and deduplication for peptide-literature evidence records. The package accepts identifiers and curator-supplied labels, produces machine-readable provenance, and keeps the evidence boundary explicit.

What it does

• normalizes DOI, PMID, PMCID, URL and local fixture identifiers;
• deduplicates equivalent identifiers without deleting the original IDs from the duplicate report;
• preserves optional title, year, evidence tier, tags, source and notes;
• computes an input SHA-256 hash for a reproducible curation pass;
• reads JSON or a documented CSV shape and emits a JSON evidence map.

Install and use

npm install peptide-evidence-map
npx peptide-evidence-map ingest \
  --file examples/evidence.synthetic.json \
  --search-date 2026-08-11 \
  --json

The example file is explicitly synthetic. A real project should keep the original search strategy, source export, access date, identifier list and curator decisions next to the generated map.

Scientific boundary

This package does not download papers, copy abstracts, judge study quality, infer efficacy, resolve conflicting clinical claims or turn preclinical evidence into medical advice. evidenceTier is a transparent curator label, not an automated assessment. The tool is research-use-only and educational software. Péptido México is a commercial supplier of research materials; that relationship is disclosed in CITATION.cff.

Development

npm test
npm run pack:check

Fixtures contain no customer data and no real literature claims. Cite the exact package release and the source records used in your own analysis. The latest suite release is archived on Zenodo.