GWAS Catalog: Data Search, API & Summary Stats

Q: What licensing terms govern data reuse from the GWAS Catalog?

Summary statistics are released under the Creative Commons Zero (CC0) public domain dedication unless explicitly noted. Visualisations and diagrams are available under CC-BY-4.0, and software tools are distributed under the Apache 2.0 license.

Q: Can researchers submit unpublished GWAS summary statistics?

Yes. The GWAS & PGS Catalog Deposition Services accept summary statistics for preprints and unpublished human genome-wide studies, labelling them as uncurated until formal peer-reviewed publication.

The GWAS Catalog (the NHGRI-EBI Catalog of published genome-wide association studies) is the primary open-access repository for human genome-wide association data. Established to organize expanding genomic literature, the platform provides systematic curation of published genetic associations alongside a public archive of full genome-wide summary statistics. By standardizing trait descriptions and mapping variants, the catalog enables researchers to evaluate associations between human phenotypes and single nucleotide polymorphisms without parsing heterogeneous supplementary tables.

Maintained as an ELIXIR Core Data Resource and Global Core Biodata Resource, the database adheres to FAIR data principles, supporting geneticists, bioinformaticians, and epidemiologists in meta-analyses, functional genomics, and causal inference.

What it is

The GWAS Catalog is a curated, searchable knowledgebase of human genome-wide association studies. When researchers conduct a GWAS, they test millions of variants across study cohorts to identify statistical associations with diseases or quantitative traits. The catalog organizes these findings into two core tiers:

Curated Top Associations: Expert curators manually extract significant variant-trait associations, mapping reported phenotypes to standard ontologies and recording coordinates, risk alleles, effect sizes, and p-values.
Full Genome-Wide Summary Statistics: A public repository of complete summary statistics files deposited by authors or collected by curators, containing association values for all analysed variants across the genome.

The platform also provides the GWAS Diagram, an interactive karyotype visualization mapping trait associations across human chromosomes.

Who runs it and what it covers

The GWAS Catalog was founded in 2008 by the National Human Genome Research Institute (NHGRI). Since 2010, it has operated as a joint collaboration between the European Molecular Biology Laboratory – European Bioinformatics Institute (EMBL-EBI) and NHGRI. The service is hosted by EMBL-EBI in Hinxton, UK. Scientific leadership is shared between Helen Parkinson at EMBL-EBI and Mike Inouye at the University of Cambridge, supported by dedicated curation and engineering teams, with guidance from an independent Scientific Advisory Board.

Primary funding is awarded under NHGRI Grant Number 1U24HG012542-01, delivered with NCBI and Cambridge, with additional support from Open Targets.

Coverage and Inclusion Criteria

Literature triage runs weekly using LitSuggest, an NCBI machine learning system, followed by bi-weekly releases of curated studies. Inclusion is governed by specific eligibility rules:

Human Germline Focus: The catalog exclusively indexes studies of the human genome. Non-human studies and somatic mutation analyses (such as tumour sequencing) are excluded.
Variant Density: Eligible studies must analyse at least 100,000 pre-quality-control variants across the genome. Candidate-gene studies and custom arrays lacking a genome-wide backbone are excluded.
Significance Cutoff: Curators extract top variant-trait associations achieving p < 1 × 10⁻⁵, recording the most significant variant per locus. This captures suggestive signals while allowing researchers to filter by the stricter genome-wide significance threshold of p < 5 × 10⁻⁸.
Genotyping Technologies: Array-based GWAS, sequencing-based studies (seqGWAS), and large targeted arrays (e.g., Metabochip, Immunochip) exceeding 100,000 variants are included.
Publication Status: The repository indexes peer-reviewed papers from PubMed and accepts author submissions of preprints and unpublished summary statistics, labelled as uncurated until journal publication.

How to search or use it

Researchers access the GWAS Catalog through its web portal, bulk downloads, and programmatic interfaces across six operational steps:

Access the Web Portal Interface: Navigate to ebi.ac.uk/gwas. The search bar supports queries by trait name, gene symbol, dbSNP variant identifier (rsID), genomic coordinate region, PubMed ID, or study accession (GCST).
Filter by Standardized Phenotypic Traits: Traits are mapped to the Experimental Factor Ontology (EFO) and MONDO disease terms. Querying allows toggling child trait inclusion (show_child_traits=true by default on trait pages).
Inspect Curated Associations and Cohorts: Results tables present reported traits, mapped genes, risk allele frequencies, p-values, odds ratios or beta coefficients, and standardized ancestry descriptors.
Download Full Summary Statistics: Locate the study accession (e.g., GCST007240) to access summary statistics hosted on the EMBL-EBI FTP server. Complete summary statistics are not delivered via the REST API and must be retrieved as flat files. These files provide core inputs for secondary analyses like Mendelian randomisation studies.
Query Programmatically via REST API v2: Submit HTTP GET requests to ebi.ac.uk/gwas/rest/api/ (v1 was retired in August 2026). Pagination defaults to 20 records. Useful parameters include efo_trait, efo_id, and extended_geneset. Rate limits allow up to 100 requests per minute and 15 requests per second per IP address (excess calls return HTTP 429). Set a descriptive User-Agent header.
Deposit Data via Globus File Transfer: Authors submit datasets through the GWAS & PGS Catalog Deposition Services by connecting a Globus account for file upload, completing the study metadata template, and receiving study accessions.

Strengths and limits

Evaluating these characteristics assists investigators in choosing when to employ the platform as their primary data repository.

Strengths

Standardized Curation: Expert curators standardize traits to EFO ontologies and map variants to uniform assemblies, resolving inconsistencies in published tables.
Open Data Licensing: Summary statistics are released under Creative Commons Zero (CC0) dedication unless otherwise noted. Diagrams are available under CC-BY-4.0, and software tools under the Apache 2.0 license.
Stable Identifiers: Immutable GCST accession numbers ensure data provenance and reproducibility.
Cross-Resource Integration: Records cross-reference Ensembl, dbSNP, Europe PMC, and Open Targets.

Limits

Scope Delimitations: Restricted to human germline variation; animal models, candidate gene studies, and somatic tumour mutations are excluded.
API Delivery Constraints: REST API v2 serves curated top associations and metadata, but does not support real-time variant querying across full summary statistics files; full files require bulk FTP download.
Curation Lag: Manual curation and bi-weekly release cycles introduce a brief lag between paper publication and appearance of curated records.

Alternatives

Researchers frequently use complementary genetics platforms tailored to specific analytical workflows:

Platform	Primary Focus	Data Scope	Access Mechanism	Licensing Model
GWAS Catalog	Global index of published GWAS and summary statistics	Global literature, ≥100k variants, human germline, top hits and summary stats	Web portal, REST API v2, Globus, FTP files	CC0 for summary statistics; diagrams CC-BY-4.0; code Apache 2.0
Open Targets Genetics	Therapeutic target identification and variant-to-gene prioritization	GWAS Catalog and UK Biobank integrated with functional genomics and V2G scores	Web portal, GraphQL API, BigQuery, GCS bulk files	EMBL-EBI open data terms; Apache 2.0 software
IEU OpenGWAS	Summary statistics formatted for Mendelian randomization	Tens of thousands of curated GWAS datasets structured for MR-Base pipelines	Web portal and REST API with API token	Academic open access; source dataset licenses apply
GWAS Atlas	Cross-trait genetic architecture and genetic correlation mapping	Curated GWAS datasets with phenome-wide correlations and pleiotropy tools	Web interface, Atlas viewer, TSV downloads	Academic open access (Beijing Institute of Genomics / CAS)
UK Biobank Showcase	Single prospective cohort phenotyping and genomics	Genotypes, exomes, and whole genomes linked to EHRs for 500,000 UK participants	Restricted application via Access Management System	Restricted access agreement with cost-recovery fee
FinnGen	Isolated population cohort linking genomics with health registries	Genomic data from >500,000 Finnish participants linked to national registries	Public web browser for summary results; individual data restricted	Public browser; individual data restricted to consortium partners

FAQ

How do I cite the GWAS Catalog in academic publications?

Reference the primary Nucleic Acids Research update paper: Cerezo M, Sollis E, Ji Y, et al. (2025). The NHGRI-EBI GWAS Catalog. Nucleic Acids Research, 53(D1):D998–D1005 (doi:10.1093/nar/gkae1070).

How should downloaded summary statistics be cited?

Authors should cite the GWAS Catalog study accession ID (e.g., GCST007240), the catalog reference (Cerezo et al., 2025), the retrieval date, and the primary publication that originally generated the dataset.

Does the GWAS Catalog REST API provide access to full genome-wide summary statistics?

No. REST API v2 provides curated top associations and metadata, but full genome-wide summary statistics are not served via the API (the earlier summary statistics API was retired in 2026). Complete files must be downloaded directly via study pages or the EMBL-EBI FTP server.

What is the p-value threshold for inclusion of top associations in the catalog?

Curators index top variant-trait associations achieving p < 1 × 10⁻⁵, recording the most significant single nucleotide polymorphism per independent locus.

What licensing terms govern data reuse from the GWAS Catalog?