GWAS Catalog MCP Server

What is GWAS Catalog MCP Server?

This MCP server provides a programmatic interface to the GWAS Catalog REST API, enabling access to GWAS study, variant, trait, and association data. It automatically handles large result sets by offering both in-memory results and file-based storage options.

How to use GWAS Catalog MCP Server?

Install dependencies with uv sync, activate the virtual environment (. .venv/bin/activate), then run the server with uv run server.py. The server exposes a single tool, GWAS_catalog, with multiple endpoints for querying different data types. All endpoints accept common parameters like max_items_in_memory, output_dir, and force_to_file.

Key features of GWAS Catalog MCP Server

• Provides programmatic access to GWAS Catalog REST API
• Automatic handling of large result sets with file storage
• Queries studies, variants, traits, and associations
• Supports genome-wide significance filtering (p ≤ 5e-8)
• Configurable memory limit and output directory
• Removes _links fields from API responses by default

Use cases of GWAS Catalog MCP Server

• Retrieve study details and associations for a given study identifier
• Search variants in a genomic region and filter by trait
• Fetch all associations for a specific variant (with significance flag)
• Obtain studies and associations linked to a trait (EFO ID)
• Rank top variants associated with a trait

FAQ from GWAS Catalog MCP Server

What happens when results exceed the memory limit?

A subset of results is returned in the items field, is_complete is set to false, and the complete dataset is automatically saved to a file specified in the output_file field.

What dependencies are required to run the server?

The server requires uv, mcp[cli], fastmcp, and requests. Install them with uv sync.

Can I force all results to be written to a file?

Yes. Set the force_to_file parameter to true to always save results to a file regardless of size. Alternatively, use force_no_file to never write to file.

Which endpoints use the GWAS Summary Statistics API instead of the main REST API?

The get_region-trait-associations endpoint and other endpoints marked as "uses GWAS Summary Statistics API" access https://www.ebi.ac.uk/gwas/summary-statistics/api.

Does the server filter results for genome-wide significance?

For endpoints that process p-values (e.g., get_associations_from_variant), only results with is_gwas_significant: true (p ≤ 5e-8) are returned. The metadata includes significant_items count.