Genomics · Database
PGS Catalog
An open database of published polygenic scores and evaluation metadata for genomic research.
What PGS Catalog is
The Polygenic Score (PGS) Catalog is an open database of published polygenic scores alongside the metadata required for their evaluation and accurate application. Developed and operated by EMBL-EBI and the University of Cambridge, with funding support from the NIH, the resource addresses the need for a consistent, centralised repository for scores that predict phenotype predisposition.
A polygenic score compiles the effects of hundreds to millions of genetic variants—typically single-nucleotide polymorphisms (SNPs)—into a single value. These are calculated as a weighted sum of allele dosages multiplied by effect sizes derived from genome-wide association studies (GWAS). Polygenic scores may also be termed genetic scores, genomic risk scores (GRS), or polygenic risk scores (PRS) when used to predict discrete disease outcomes.
The catalogue provides researchers with consistent annotations, scoring files, developmental background, and evaluations of predictive performance across hundreds of traits and publications, enabling the reproduction and calculation of published scores on external genomic datasets.
How it is organised
Each entry in the PGS Catalog is curated to include the underlying scoring file containing genetic variants, effect alleles, and corresponding weights. It also captures the metadata outlining how the score was developed, applied, and evaluated for predictive capability.
Data can be accessed and retrieved through multiple interfaces, including web search, bulk downloads from the website and FTP, programmatic access via a REST API, and a command-line interface using the pgscatalog_utils Python package. The catalogued scores can be computed alongside custom scores on user data using the pgsc_calc reproducible workflow tool.
Key facts
- Polygenic scores
- 6,991
- Traits covered
- 813
- Publications included
- 836
- Developer organisations
- EMBL-EBI and University of Cambridge
- Associated computation software
- pgsc_calc
- CLI download utility
- pgscatalog_utils Python package
- Programmatic interface
- REST API
- Bulk access methods
- Website and FTP
What you can use it for
- Polygenic score calculation
- Download variant weights and use the pgsc_calc pipeline to calculate published polygenic scores on independent genomic data.
- Predictive performance evaluation
- Review curated performance evaluations and development metadata to compare how different scores predict specific diseases or traits.
- Custom and catalogued score workflows
- Run reproducible score calculations combining both catalogued scores and newly generated custom polygenic scores via pgsc_calc.
- Bulk genomic data retrieval
- Retrieve complete sets of scoring files and associated metadata programmatically via the REST API or FTP for large-scale analysis.
Query it here
The research agent queries PGS Catalog directly — no account, no API key — and every answer links back to the record it came from. Ask in plain language and it writes the query.
Strengths and limits
Good at
- Provides consistently annotated scoring files detailing variants, effect alleles, and weights for thousands of scores.
- Captures explicit metadata detailing both score development and evaluations of predictive performance.
- Supplies companion command-line and workflow software (pgsc_calc and pgscatalog_utils) to facilitate reproducible score calculation.
Watch out for
- Limited to published scores, preprints, or submitted pre-release studies meeting catalog inclusion criteria.
- Scores are dependent on the effect size estimates and variant sets derived from underlying genome-wide association studies.
Questions
Can I search PGS Catalog without writing code?
Yes. The research agent queries PGS Catalog directly, with no account or API key, and cites the record behind every answer. Ask in plain language — for a gene, a protein, a compound or a study — and it builds the query for you.
Can I reuse data from PGS Catalog?
EMBL-EBI Terms of Use; each score carries its authors' own licence. Report the licence on the score record before reusing its weights; cite the score's publication. Terms: https://www.ebi.ac.uk/about/terms-of-use/
What is the difference between a PGS, PRS, and GRS?
These terms refer to the same general concept of aggregating multiple genetic variant effects using a weighted sum. When predicting discrete phenotypes such as disease status, they are frequently termed polygenic risk scores (PRS) or genomic risk scores (GRS).
How can I calculate scores from the catalog on my own data?
You can obtain the scoring files containing variant weights and use the pgsc_calc software, a reproducible workflow designed to compute both Catalog-derived and custom scores.
How can I access the catalogue programmatically?
The database offers a REST API for programmatic access to metadata, as well as the pgscatalog_utils Python package for command-line downloads.
Written from PGS Catalog's own website and reference material. Statements citing a figure that does not appear there are removed before publication. Licence and running details come from our registry.