SynDRA: Structure-Anchored Synonym and Cross-Reference Hub for Drug Repurposing Alignment (v1.0.2026Q3)

SynDRA (Synchronized Drug Repurposing Atlas) is a structure-anchored compound hub that integrates synonyms and cross-references from nine biomedical resources (LINCS 2020 CMap, TTD, PRISM, DrugCentral, DrugBank, PubChem, ChEMBL, UniChem, RxNorm), plus Wikidata and FDA Orange Book extensions, into 3,113,733 canonical compound nodes backed by 12.2 million synonyms and 20.3 million cross-references spanning 25 external identifier types. This dataset contains the redistributable build (CC BY-SA 4.0): three tables (compounds, synonyms, cross-references) in both Apache Parquet and CSV formats, suitable for direct pipeline integration, plus an RDF Turtle export (syndra.ttl) in which every compound is minted as a dereferenceable Linked Data URI (schema.org/SKOS/OWL/DCTerms/CHEMINF vocabularies, https://syndrabio.org/resource/{sid}). The interactive web portal queries a SQLite database hosted on Cloudflare R2 via HTTP Range requests (sql.js-httpvfs) — no full-index download required. Source code and build pipeline: https://github.com/hidelab/SynDRA Interactive web portal: https://syndrabio.org/ REST/GraphQL API: https://api.syndrabio.org