Scientific Claim Registry (SCR): A Framework for Persistent, Versioned, Expert-Curated Biomedical Claims

Modern science assigns persistent identifiers to nearly every artifact it produces — papers (DOI), researchers (ORCID), clinical trials (registration numbers), genetic sequences (accession numbers) — yet the scientific claim itself has no identity. Claims remain trapped inside publications: un-addressable, un-versionable, and impossible to track as evidence accumulates. This work proposes the Scientific Claim Registry (SCR): a public, versioned, expert-curated registry that assigns each biomedical claim a persistent identifier, a canonical statement with explicit context (population, condition, exposure, comparator, outcome, methodological scope), links to supporting and contradicting evidence, an evidence-certainty rating reusing GRADE, and a complete revision history. The registry is operated by the BIO (Biological Intelligence Observatory) method — its governance and maintenance layer. SCR is presented as a testable hypothesis: whether a claim-centric, versioned, expert-curated registry can maintain current biomedical knowledge better than publication-centric synthesis. Lipedema is proposed as the first proof-of-concept domain; the framework is disease-agnostic. The work explicitly builds on and acknowledges prior art including nanopublications, micropublications, Wikidata, CIViC, ClinGen, PharmGKB/CPIC, Open Targets, Cochrane Living Systematic Reviews, GRADE and MAGICapp; its proposed contribution is adoption, governance, and domain execution, not the claim primitive itself. This document establishes the conceptual foundation and priority date of the SCR/BIO framework. Future technical specifications, implementations, and disease-specific deployments will build upon the principles described herein.