Written and maintained by CASRAI Editorial Board
Last updated
NCBI Virus is a web portal from the National Center for Biotechnology Information (NCBI) at the US National Library of Medicine for finding, filtering, aligning and downloading viral nucleotide and protein sequences. This guide is for researchers who need a defined set of viral sequences, such as all human host SARS-CoV-2 genomes from one region, and want to know which records the portal returns, how to export them, and where its limits lie. It is based on the portal’s own Help and About pages, read on 2026-11-07.
What NCBI Virus is
NCBI Virus is a front end over public viral sequence data held at NCBI. It combines a search form, a results table with filters, built-in BLAST, a quick alignment and tree tool, and a three-step download wizard. It is a search and retrieval interface, not a separate archive: the sequences come from GenBank and related NCBI resources.
Who runs it and what it covers
The About page credits the National Library of Medicine (US), National Center for Biotechnology Information. Its publications list includes the 2017 Nucleic Acids Research article “Virus Variation Resource – improved response to emergent viral outbreaks” and the 2008 Journal of Virology article “The Influenza Virus Resource at NCBI”. The portal’s menus are Find Data, Help, How to Participate, Submit Sequences (separate routes for SARS-CoV-2 and for all other viruses), Contact Us, and an Account menu where the sign-in sits.
The three record types: GenBank, RefSeq and Assembly
The Help page describes a data model that explains most of what you see in the results table.
- GenBank records include all sequences submitted to the International Nucleotide Sequence Database Collaboration (INSDC), including those submitted through SRA.
- RefSeq records are reference sequences, usually based on exemplar isolates recognised by the International Committee on Taxonomy of Viruses (ICTV). There is one RefSeq per genome segment, with NC_, NP_ or YP_ accessions. By default they are shown at the top of the results table, which is why a search often opens with a handful of reference rows before the bulk of GenBank records. See the guide to NCBI Reference Sequence (RefSeq) for how RefSeq differs from GenBank.
- Assembly records include RefSeq genome assemblies and groups of segments belonging to one segmented virus (GCA accessions), produced by the NCBI Virus Segmented Genome Grouping Pipeline. Grouping applies only to influenza A, B and C (eight segments for A and B, seven for C). Segments are matched on species, collection date, collection location, isolate name and host.
Virus names are validated against NCBI Taxonomy, which follows ICTV classification. The Help page notes that taxonomy updates can lag behind ICTV decisions.
How to search and download viral sequences
The steps below use the interface terms in the official Help page.
- Search by Virus. On the home page, enter a virus name or NCBI taxonomy ID. Quick links and Popular Searches offer shortcuts for human viruses, SARS-CoV-2 and influenza.
- Read the Results Table. Each row is a sequence record. RefSeq rows appear first by default.
- Apply filters. Filters are grouped as Identifiers and Classification, Sequence Quality, Location and Source, Dates, Genome Organization, and Sample Type. Examples include Virus/Taxonomy, Accession, GenBank/RefSeq, Host, Geographic Region, Collection Date, Segment and Vaccine Strain. Multiple filters combine with AND. “Reset All” clears them.
- Select rows or keep all results. Use “Download Results” for ticked rows or “Download All Results” for the whole filtered set.
- Choose the data type (download step 1). Options are Nucleotide, Protein or CDS FASTA, a list of accessions, or the Results Table as CSV or XML.
- Choose the records (step 2). Take the selected records, all records, or a randomized subset of up to 2,000, which can be stratified by Country, Collection Year, Release Year or Host.
- Set FASTA titles (step 3). The default title is the accession plus the GenBank title. “Build custom sequence title” lets you compose your own header from metadata fields.
Other tools in the portal
- Search by Sequence runs BLAST against NCBI Virus: BLASTn for nucleotide or BLASTp for protein queries, one query at a time, with a Betacoronavirus DB option.
- Outbreak Sequence Statistics shows submissions over the last 10 weeks for H5N1, MPXV, measles and SARS-CoV-2, with an optional BioProject filter. For BioProject context see NCBI BioProject.
- Align uses MUSCLE on selected rows and exports FASTA, PDV or SVG.
- Build Phylogenetic Tree exports ASN, Newick, Nexus or PDF.
Strengths and limits
Strengths
- One interface for GenBank, RefSeq and assembly records, with RefSeq references surfaced first.
- Filters on host, geography, collection date, sequence quality and vaccine strain, rather than only on accession or name.
- A download wizard that produces FASTA, accession lists or metadata tables, including a stratified random subset of up to 2,000 records.
Limits stated in the Help page
- Alignments are for quick visualisation and preliminary work, not publication-quality analysis.
- The tree is built from BLAST similarity scores and is not a true phylogenetic method, so it is not definitive evidence of relatedness.
- Nucleotide completeness is inferred from the GenBank record, not always taken from what the submitter declared, so check completeness for any analysis that depends on full-length genomes.
- Taxonomy can lag ICTV classification.
- Very large sets should be fetched from the NCBI FTP site or with NCBI Datasets rather than through the browser wizard.
Alternatives and companion routes
| Route | What the NCBI Virus Help page says about it |
|---|---|
| NCBI Datasets | Command-line and programmatic access to data; the Help page points here for large or scripted retrieval. |
| NCBI FTP site | Full metadata in CSV and the complete nucleotide FASTA, for bulk download. |
| GenBank | The underlying INSDC sequence archive whose records NCBI Virus indexes. |
| RefSeq | Curated reference sequences, shown first in NCBI Virus. |
FAQ
Does NCBI Virus have an API?
The Help page describes no dedicated NCBI Virus API. For programmatic access it points to NCBI Datasets and to the NCBI FTP site (checked 2026-11-07).
How do I download SARS-CoV-2 genomes from NCBI Virus?
Choose the SARS-CoV-2 quick link or search by name, filter (for example by Host, Geographic Region and Collection Date), then use Download All Results and pick Nucleotide FASTA. Use the Betacoronavirus DB option under Search by Sequence if you want to BLAST a query against that database.
Is NCBI Virus the same as GenBank?
No. GenBank is the sequence archive; NCBI Virus is a search, filter and download interface that also surfaces RefSeq and assembly records.
Why do RefSeq records appear first?
By default the Results Table lists RefSeq reference records at the top, ahead of GenBank records.
Can I use the built-in tree in a publication?
The Help page says the tree is based on BLAST similarity scores, is not a true phylogenetic method and is not definitive evidence, so build publication trees with dedicated phylogenetic software.
Sources
- NCBI Virus Help, https://www.ncbi.nlm.nih.gov/labs/virus/vssi/docs/help/ (read 2026-11-07)
- NCBI Virus About, https://www.ncbi.nlm.nih.gov/labs/virus/vssi/docs/about/ (read 2026-11-07)
- NCBI Virus portal, https://www.ncbi.nlm.nih.gov/labs/virus/vssi/ (read 2026-11-07)
