Compound Databases

Compound databases support drug-discovery research by organizing chemical structures, identifiers, bioactivity values, targets, annotations, purchasability, and literature-derived evidence into searchable resources for virtual screening, target prediction, repurposing, and hit prioritization. General chemical databases such as PubChem, ChEMBL, DrugBank, BindingDB, and ZINC differ in scope: PubChem emphasizes broad chemical and assay data, ChEMBL and BindingDB emphasize bioactivity and target relationships, DrugBank emphasizes approved and investigational drugs, and ZINC emphasizes purchasable docking-ready compounds. Natural-product and traditional-medicine databases add species source, phytochemical, and ethnopharmacology information, but reviews show that database accessibility, redundancy, inconsistent structure annotation, and uneven curation remain major limitations. Unresolved questions include how to harmonize identifiers across databases, how to remove duplicate or erroneous st