refseq

RefSeq is the NCBI Reference Sequence database, a curated, non-redundant collection of DNA, RNA, and protein sequences that serves as a standard reference for modeling genomes and gene products. Maintained by the National Center for Biotechnology Information, RefSeq aggregates sequences from multiple sources and assigns stable accessions and consistent annotations to facilitate comparison and interpretation across studies. The collection covers a broad range of organisms, from model species to clinically relevant pathogens, and includes genomic DNA sequences, curated transcript sequences, and protein sequences. It also contains computationally generated or predicted sequences for organisms or genes lacking curated entries, clearly labeled as such.

Access to RefSeq is available through NCBI’s Entrez system and FTP sites, with new releases and versioned updates that track changes to sequences and annotations. Accession prefixes indicate the sequence type, for example NC_ for complete genomic sequences, NM_ for mRNA, NP_ for protein, and NR_ for non-coding RNA; prefixes such as XM_, XP_, XR_ denote computationally predicted alternatives. RefSeq documents alignments, coding status, gene structure, and genomic context, often linking to related RefSeq records and external resources.

Applications include use as standard references in genome annotation, comparative genomics, expression studies, and clinical variant interpretation, as well as as an anchor for other NCBI resources like Gene, ClinVar, and dbSNP. RefSeq is continually updated to incorporate new data and improve annotations while preserving historical versions for reproducibility.