
OpenTargets MCP Server
Unofficial Model Context Protocol server for accessing Open Targets platform data for gene-drug-disease associations research.
README
Unofficial Open Targets MCP Server 🧬
Unofficial Model Context Protocol server for accessing Open Targets platform data for gene-drug-disease associations research.
Developed by Augmented Nature
✅ Verified Features
All 6 tools working with live Open Targets API data:
- 🎯 Target Search - Gene symbols, names, descriptions (BRCA1, TP53, etc.)
- 🦠 Disease Search - Names, synonyms, descriptions (cancer, diabetes, etc.)
- 🔗 Target-Disease Associations - Evidence scores from 20+ databases
- 📊 Disease Target Summaries - Prioritized therapeutic targets
- 🧬 Target Details - Comprehensive gene/protein information
- 🎭 Disease Details - Complete disease profiles with ontologies
🚀 Quick Start
# Install and build
npm install
npm run build
# Run the server
node build/index.js
📋 MCP Client Configuration
Claude Desktop
{
"mcpServers": {
"opentargets-server": {
"command": "node",
"args": ["/path/to/opentargets-server/build/index.js"]
}
}
}
Other MCP Clients
node /path/to/opentargets-server/build/index.js
🛠️ Available Tools
🎯 search_targets
Search therapeutic targets by gene symbol, name, or description
{
"name": "search_targets",
"arguments": {
"query": "BRCA1", // Gene symbol, name, or description
"size": 10 // Optional: 1-500 results (default: 25)
}
}
Example Results:
- BRCA1 (ENSG00000012048) - BRCA1 DNA repair associated
- BRCA2 (ENSG00000139618) - BRCA2 DNA repair associated
- BRIP1 (ENSG00000136492) - BRCA1 interacting DNA helicase 1
🦠 search_diseases
Search diseases by name, synonym, or description
{
"name": "search_diseases",
"arguments": {
"query": "breast cancer", // Disease name, synonym, or description
"size": 10 // Optional: 1-500 results (default: 25)
}
}
🔗 get_target_disease_associations
Get target-disease associations with evidence scores
{
"name": "get_target_disease_associations",
"arguments": {
"targetId": "ENSG00000012048", // Target Ensembl ID
"size": 10 // Optional: 1-500 results
}
}
OR
{
"name": "get_target_disease_associations",
"arguments": {
"diseaseId": "EFO_0000305", // Disease EFO ID
"size": 10 // Optional: 1-500 results
}
}
📊 get_disease_targets_summary
Get prioritized targets associated with a disease
{
"name": "get_disease_targets_summary",
"arguments": {
"diseaseId": "EFO_0000305", // Disease EFO ID (required)
"size": 20 // Optional: 1-500 targets (default: 50)
}
}
🧬 get_target_details
Get comprehensive target information
{
"name": "get_target_details",
"arguments": {
"id": "ENSG00000012048" // Target Ensembl gene ID
}
}
🎭 get_disease_details
Get comprehensive disease information
{
"name": "get_disease_details",
"arguments": {
"id": "EFO_0000305" // Disease EFO ID
}
}
📚 Resource Templates
Access Open Targets data through standardized URIs:
opentargets://target/{ensemblId}
- Complete target informationopentargets://disease/{efoId}
- Complete disease informationopentargets://drug/{chemblId}
- Drug informationopentargets://association/{targetId}/{diseaseId}
- Association evidenceopentargets://search/{query}
- Search results
🧪 Real-World Examples
Cancer Research Workflow
# 1. Search for cancer-related targets
{"name": "search_targets", "arguments": {"query": "oncogene", "size": 10}}
# 2. Get detailed info for specific target
{"name": "get_target_details", "arguments": {"id": "ENSG00000012048"}}
# 3. Find all diseases associated with BRCA1
{"name": "get_target_disease_associations", "arguments": {"targetId": "ENSG00000012048"}}
# 4. Get top targets for breast cancer
{"name": "get_disease_targets_summary", "arguments": {"diseaseId": "EFO_0000305", "size": 20}}
Drug Discovery Pipeline
# 1. Search for Alzheimer's disease
{"name": "search_diseases", "arguments": {"query": "Alzheimer", "size": 5}}
# 2. Get disease details
{"name": "get_disease_details", "arguments": {"id": "EFO_0000249"}}
# 3. Find prioritized therapeutic targets
{"name": "get_disease_targets_summary", "arguments": {"diseaseId": "EFO_0000249", "size": 30}}
🔬 Data Sources & Standards
Open Targets integrates 20+ databases:
- ChEMBL - Drug & compound data
- Ensembl - Gene & protein annotations
- EFO - Experimental Factor Ontology
- ClinVar - Clinical variant data
- GWAS Catalog - Genome-wide association studies
- UniProt - Protein sequences & functions
- Reactome - Biological pathways
- And many more...
Standardized Identifiers:
- Targets: Ensembl gene IDs (e.g., ENSG00000012048)
- Diseases: EFO IDs (e.g., EFO_0000305)
- Drugs: ChEMBL IDs (e.g., CHEMBL1234)
🏗️ Architecture
- TypeScript implementation with robust type safety
- GraphQL queries for efficient data retrieval
- MCP Protocol compliant JSON-RPC communication
- Error Handling with comprehensive validation
- Production Ready with 30s timeouts and proper logging
📊 API Information
- Base URL:
https://api.platform.opentargets.org/api/v4/graphql
- Version: Open Targets v25.0.1
- Rate Limits: Generous for research use
- Authentication: None required
- Format: GraphQL queries, JSON responses
🤝 Contributing
This server is developed and maintained by Augmented Nature. For enhancements:
- Fork the repository
- Make your changes
- Submit a pull request
Support
For issues with:
- MCP Server: Check the server logs and error outputs
- Open Targets API: Visit platform.opentargets.org
- GraphQL Queries: Use the Open Targets GraphQL browser
Citation
If you use this project in your research or publications, please cite it as follows:
author = {Moudather Chelbi},
title = {OpenTargets MCP Server},
year = {2025},
howpublished = {https://github.com/Augmented-Nature/OpenTargets-MCP-Server},
note = {Accessed: 2025-06-29}
推荐服务器

Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。

VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。

e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。