opentargets-mcp

Overview

This MCP server bridges the gap between AI clients and the Open Targets Platform by exposing its GraphQL API through a comprehensive set of tools. It covers target analysis, disease research, drug discovery, evidence mining, variant analysis, and study exploration. The implementation focuses on practical biomedical workflows with features like strict ID resolution, typed parameter handling, and efficient data filtering options.

Try asking AI

After installing, here are 5 things you can ask your AI assistant:

• get_target_info

  Core target identity record (Ensembl IDs, synonyms, genomic coordinates)

  Try:Get information about the EGFR targetTry:Show me details for the TP53 gene
• get_disease_info

  Disease/EFO summary with therapeutic area context

  Try:Get information about schizophreniaTry:Show me details for type 2 diabetes
• get_drug_info

  ChEMBL-backed drug profile and mechanism data

  Try:Get information about osimertinibTry:Show me details for aspirin
• search_entities

  Unified entity search with synonym handling

  Try:Search for targets related to Alzheimer'sTry:Find drugs for breast cancer
• get_target_associated_diseases

  High-confidence target-disease links with scores

  Try:What diseases is BRCA1 associated with?Try:Show diseases linked to EGFR
• get_target_disease_evidence

  Evidence details across genetics, expression, and literature

  Try:Get evidence for TP53 and lung cancerTry:Show evidence linking EGFR and glioblastoma
• get_drug_repurposing_candidates

  Multi-hop disease -> target -> drug candidate prioritization

  Try:Find repurposing candidates for inflammatory bowel diseaseTry:What drugs could be repurposed for Alzheimer's?
• get_variant_info

  Information about genetic variants

  Try:Get details about the rs12345 variantTry:Show information about rs6764265
• get_study_info

  Information about GWAS studies

  Try:Get details about study ABC123Try:Show information for GWAS study UK Biobank 1234
• graphql_query

  Execute a raw GraphQL query

  Try:Run a GraphQL query for target expression dataTry:Get all known drug-target interactions via GraphQL
• map_ids

  Map between different identifier systems

  Try:Convert EGFR to Ensembl IDTry:Map ChEMBL123 to drug name
• get_drug_adverse_events

  Adverse event information for a drug

  Try:What adverse events are associated with osimertinib?Try:Show adverse events for metformin

Comparable tools

Installation

pip install mcpm
mcpm install opentargets

uvx --from git+https://github.com/nickzren/opentargets-mcp opentargets-mcp

git clone https://github.com/nickzren/opentargets-mcp
cd opentargets-mcp
pip install uv
uv sync
uv run python -m opentargets_mcp.server

{
  "mcpServers": {
    "opentargets": {
      "command": "uv",
      "args": ["run", "python", "-m", "opentargets_mcp.server", "--transport", "stdio"]
    }
  }
}

FAQ

What transport modes are supported?

The server supports stdio (default), SSE, and HTTP transports powered by FastMCP.

Can I filter the data returned by the tools?

Yes, many core tools accept an optional 'fields' parameter to filter the response payload and reduce token overhead.

Compare opentargets-mcp with