Overview
The National Center for Biotechnology Information (NCBI) at the U.S. National Library of Medicine provides the world’s most comprehensive suite of molecular biology databases. GenBank, the primary nucleotide sequence repository, accepts direct submissions from researchers and participates in the International Nucleotide Sequence Database Collaboration (INSDC) alongside EMBL-EBI and DDBJ. RefSeq provides curated, non-redundant reference sequences for genomes, transcripts, and proteins. The Sequence Read Archive (SRA) stores raw sequencing reads from high-throughput platforms. Together, these resources form the backbone of public molecular data access.
Key Concepts
GenBank records include the raw sequence, biological annotations, and bibliographic metadata. Accession numbers (e.g., NM_001234) provide stable identifiers for citation. RefSeq differs from GenBank in its curation process: RefSeq records are manually reviewed and maintained, offering higher reliability for reference purposes. SRA archives unprocessed sequencing reads in a compressed format, storing both the read data and quality scores. The Entrez search system provides unified cross-database querying across all NCBI resources, enabling retrieval of related records from different databases through a single interface.
Applications
NCBI databases are used daily by researchers worldwide. DNA sequencing projects begin with BLAST searches against GenBank to identify unknown sequences. Next-generation sequencing experiments deposit raw reads in SRA and align them against RefSeq genomes for variant calling. Bacterial genetics studies rely on RefSeq genomes for comparative genomics and the identification of species-specific genes and virulence factors.
Practical Protocol
Navigating the NCBI portal starts at ncbi.nlm.nih.gov, where the search bar provides access to the Entrez cross-database search system. To find all nucleotide sequences for a given organism, enter the organism name (e.g., “Escherichia coli”) in the main search bar, then use the “Nucleotide” database filter on the results page. For more precise queries, use Entrez field syntax: “Escherichia coli[Organism] AND 16S ribosomal RNA[Title]” returns only 16S rRNA sequences from E. coli. The results list shows accession numbers, sequence lengths, and definitions. Click any accession (e.g., NR_024570) to view the full GenBank record with annotated features, coding sequences, and references. For batch retrieval, click “Send to” > “File” > “FASTA” to download all matching sequences. For a practical example, a microbiologist characterizing a novel bacterial isolate would: (1) search SRA for raw reads from closely related species to use as a reference for assembly, (2) submit the assembled contigs to BLAST against the RefSeq genome database to identify the closest sequenced relative, (3) download the RefSeq genome of that relative for gene prediction with tools like Prokka, and (4) deposit the annotated assembly in GenBank with a new accession. The entire workflow, query, retrieve, analyze, and submit, operates within the unified NCBI ecosystem, with Entrez providing seamless cross-database navigation at every step. For example, a PubMed search for a pathogen of interest links directly to Nucleotide records for its genome, to SRA for available sequencing data, and to BioProject for experimental metadata, all without leaving the NCBI portal.
