- GenBank and RefSeq are two major nucleotide sequence resources provided through the National Center for Biotechnology Information (NCBI), but they serve different purposes in biological research and bioinformatics. Because both resources contain DNA and RNA sequence information, they are sometimes treated as interchangeable. However, understanding the difference between GenBank and RefSeq is important when searching for sequences, performing BLAST analyses, annotating genomes, identifying genes, conducting comparative genomics, or selecting reference sequences for computational studies. In simple terms, GenBank is primarily an archival repository of publicly submitted sequence data, whereas RefSeq is a curated collection of reference sequences designed to provide stable, representative sequence records.
- GenBank is one of the world’s major public nucleotide sequence databases. It contains publicly available DNA sequences and associated annotations submitted by researchers, sequencing projects, and other data-generation efforts. GenBank is part of the International Nucleotide Sequence Database Collaboration (INSDC), together with the DNA Data Bank of Japan (DDBJ) and the European Nucleotide Archive (ENA). Sequence data submitted to one of these collaborating repositories are exchanged among the partners, allowing researchers around the world to access a shared international archive of nucleotide sequence information. This international framework is one reason GenBank has become such an important resource for modern molecular biology and genomics.
- The primary purpose of GenBank is archival data submission and dissemination. Researchers can submit nucleotide sequence data to GenBank, where the data become part of the public sequence record. GenBank therefore preserves a very broad representation of submitted sequence information. Records may represent individual genes, transcripts, complete genomes, partial sequences, environmental samples, genome assemblies, transcriptome assemblies, and many other types of nucleotide sequence data. GenBank also preserves information associated with these sequences, including organism information, sequence features, publications, and other annotations.
- RefSeq, in contrast, is the NCBI Reference Sequence database. It provides a curated, integrated, nonredundant set of reference sequences representing genomes, transcripts, and proteins. RefSeq was developed to provide stable reference records that can be used consistently for biological annotation and sequence analysis. Rather than attempting to preserve every submitted sequence as an archival record, RefSeq focuses on selecting and maintaining representative reference sequences.
- This difference in purpose is the most important distinction between the two resources. GenBank emphasizes archival preservation and comprehensive submission of publicly available sequence data, whereas RefSeq emphasizes curated reference sequences and biological representation. GenBank can therefore contain many sequences representing the same or closely related biological entities, while RefSeq attempts to provide a more streamlined collection of representative sequences.
- GenBank’s broad archival nature means that researchers can encounter considerable diversity in the quality, completeness, experimental origin, annotation, and biological characteristics of records. A GenBank record may contain a highly complete and carefully annotated genome, but another record may represent a partial sequence submitted for a specific research purpose. The presence of a sequence in GenBank does not by itself mean that NCBI has independently validated every biological claim associated with the sequence. This distinction is important when interpreting GenBank data.
- RefSeq applies additional curation and integration processes to create reference records. RefSeq sequences are intended to represent biological entities in a consistent manner and are associated with standardized identifiers and annotations. Depending on the type of RefSeq record and organism, curation can involve computational analysis, expert review, integration of information from multiple sources, and maintenance of stable reference representations.
- Another important difference concerns redundancy. GenBank can contain many records for the same gene or organism because different researchers may independently submit related sequences. These records are valuable because they preserve the original submitted data and provide evidence of sequence variation, geographic sampling, experimental observations, or independent sequencing efforts. RefSeq, on the other hand, is designed to reduce unnecessary redundancy and provide representative reference sequences that are easier to use in standardized analyses.
- For example, imagine that researchers from many laboratories sequence the same bacterial gene from hundreds of isolates. Those sequences can be submitted to GenBank as individual records. The collection may reveal genuine biological variation among the isolates and therefore be highly valuable for population genetics, epidemiology, phylogenetics, and evolutionary studies. A RefSeq record may instead provide a representative reference sequence for the organism or gene, depending on the available data and RefSeq’s representation criteria.
- This difference becomes especially important when using BLAST. Researchers can search GenBank-derived databases to find homologous sequences and explore the diversity of publicly available sequence data. A broad search may return many closely related records because GenBank preserves extensive submitted sequence information. RefSeq-based searches can provide a more curated and representative set of sequences, which can sometimes make the results easier to interpret when the goal is to identify a reference sequence or characterize a known gene.
- Neither resource is universally better than the other. The appropriate choice depends on the research question. If the goal is to investigate sequence diversity, locate original submitted sequences, explore naturally occurring variation, or search broadly across publicly available nucleotide data, GenBank can be particularly valuable. If the goal is to obtain a stable, curated, representative reference sequence for a gene, transcript, genome, or protein, RefSeq may be more appropriate.
- The difference is also important for genome annotation. Reference sequences can provide useful evidence for identifying genes and predicting coding regions in newly sequenced genomes. RefSeq records are widely used as reference material in annotation pipelines because their standardized representation can facilitate consistent comparison. GenBank records can also provide valuable annotation evidence, particularly when searching for experimentally determined sequences or sequences from closely related organisms.
- GenBank and RefSeq also differ in how their records are associated with the broader NCBI sequence ecosystem. GenBank records are submitted to the archival database and receive GenBank accession numbers. RefSeq records use RefSeq accession prefixes that distinguish them from GenBank accessions. Common RefSeq prefixes include NC_, NG_, NM_, NR_, NP_, and WP_, with the particular prefix indicating the type of reference sequence.
- GenBank accession numbers and RefSeq accession numbers should therefore not be confused. A GenBank accession identifies a particular submitted sequence record, whereas a RefSeq accession identifies a reference sequence maintained within the RefSeq collection. Understanding these identifiers is an important part of working with NCBI sequence resources.
- Versioning is another useful concept. Both GenBank and RefSeq records can have accession versions that help identify particular versions of sequence records. A sequence accession such as AB123456.1 identifies version 1 of a particular accession. If the sequence is changed and the database assigns a new version, the version number can change while the underlying accession remains associated with the record. Recording accession versions can therefore improve reproducibility when a sequence is used in an analysis.
- The two resources also differ in their relationship to sequence submissions. GenBank directly preserves submitted sequence information as part of its archival role. RefSeq does not simply function as a second copy of everything submitted to GenBank. Instead, RefSeq integrates information from multiple sources and produces reference records according to its own curation and representation processes. Consequently, a sequence being present in GenBank does not necessarily mean that an equivalent RefSeq record exists.
- Another important distinction is that RefSeq includes reference sequences beyond nucleotide sequences. The RefSeq collection includes reference genomes, transcripts, and proteins. This makes RefSeq particularly useful when researchers need coordinated reference representations across different molecular levels. For example, a researcher may move from a reference genomic sequence to a reference transcript and then to a reference protein while maintaining relationships among the records.
- GenBank, by contrast, is fundamentally an archival nucleotide sequence repository, although its records can contain extensive biological annotation and cross-references to other databases. Its enormous diversity makes it useful for discovering sequences that may not yet be represented by a corresponding RefSeq reference record.
- The relationship between GenBank and RefSeq can therefore be thought of as complementary rather than competitive. GenBank provides breadth, archival preservation, and access to submitted sequence diversity. RefSeq provides curated reference representations intended to support consistent biological interpretation and computational analysis. Researchers frequently use both resources together rather than choosing only one.
- For example, a researcher studying a bacterial gene might first use GenBank to retrieve sequences from many strains and geographic locations. The researcher could then compare these sequences with a corresponding RefSeq reference sequence to determine how individual isolates differ from a representative reference. In this workflow, GenBank provides diversity while RefSeq provides a useful reference point.
- A similar approach can be used in comparative genomics. A researcher may use RefSeq genomes as standardized references for comparing species while returning to GenBank to investigate additional isolates, draft genomes, environmental sequences, or individual submissions. Combining both resources can provide a broader understanding than either database alone.
- The distinction also matters when interpreting search results. A sequence identified in GenBank is not automatically equivalent to a curated reference sequence. Researchers should examine the record’s annotation, organism information, sequence completeness, publication details, accession version, and other available metadata before using it as evidence. For important analyses, the choice of reference sequence should be based on the biological question rather than simply selecting the first database result.
- GenBank’s enormous size also means that database searches can produce many redundant or highly similar sequences. This is useful when studying sequence variation but may complicate analyses in which researchers want a smaller representative dataset. RefSeq’s curated and nonredundant design can be advantageous in such situations because it provides a more manageable reference collection.
- At the same time, RefSeq should not be interpreted as a replacement for GenBank. The archival diversity preserved in GenBank is scientifically valuable. Individual submitted sequences may contain information about strains, populations, geographic locations, experimental samples, or unusual variants that would not necessarily be represented by a single reference sequence. Removing this diversity would reduce the information available to researchers studying biological variation.
- GenBank and RefSeq also have different roles in reproducible research. When reporting a sequence used in an analysis, researchers should identify the database source and provide the accession number, preferably including the accession version when appropriate. This allows other researchers to determine exactly which sequence was analyzed. If a reference genome or transcript was obtained from RefSeq, the corresponding RefSeq accession should be reported rather than describing it simply as a sequence obtained from NCBI.
- The distinction between the two resources becomes increasingly important as genomic datasets continue to expand. Modern sequencing projects generate enormous numbers of nucleotide sequences, assemblies, transcripts, and other data types. An archival repository such as GenBank is essential for preserving this expanding body of information, while curated reference resources such as RefSeq help researchers navigate the complexity by providing standardized reference sequences.
- It is also useful to distinguish both GenBank and RefSeq from other NCBI resources. For example, the NCBI Nucleotide database provides a search interface through which researchers can access nucleotide sequence records from several underlying sources. Similarly, NCBI’s genome and protein resources provide additional ways of accessing and analyzing sequence information. Understanding how these resources are connected can make sequence retrieval and bioinformatics workflows much more efficient.
- In practical terms, researchers can follow a simple decision-making approach. Use GenBank when you need broad access to submitted nucleotide sequence data, sequence diversity, individual isolates, original records, or archival sequence information. Consider RefSeq when you need a curated and representative reference sequence for standardized analysis, genome annotation, gene characterization, or reference-based comparison. In many research projects, the best strategy is to use both.
- A useful way to remember the difference is to think of GenBank as a comprehensive public archive and RefSeq as a curated reference collection. GenBank preserves the diversity of submitted sequence data, while RefSeq provides selected reference representations designed for consistent biological and computational use. These different purposes explain why the two databases can contain related sequences without being identical.
- Understanding GenBank versus RefSeq is particularly important for students learning bioinformatics because database choice can influence the results of sequence searches and downstream analyses. A researcher who understands the strengths and limitations of each resource can select appropriate databases for BLAST searches, sequence identification, genome annotation, comparative genomics, phylogenetics, and other applications.
- In summary, GenBank and RefSeq are complementary NCBI sequence resources with different objectives. GenBank serves primarily as an archival repository of publicly submitted nucleotide sequences and associated information, while RefSeq provides curated, integrated, and representative reference sequences. GenBank is particularly valuable for breadth and sequence diversity, whereas RefSeq is especially useful when a stable reference representation is needed. Knowing when and why to use each resource is an essential skill for anyone working with NCBI databases and modern bioinformatics.
