README.md•6.58 kB
# Unofficial Open Targets MCP Server 🧬
Unofficial Model Context Protocol server for accessing Open Targets platform data for gene-drug-disease associations research.
[](https://platform.opentargets.org/)
Developed by [Augmented Nature](https://augmentednature.ai)
## ✅ **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**
```bash
# Install and build
npm install
npm run build
# Run the server
node build/index.js
```
## 📋 **MCP Client Configuration**
### Claude Desktop
```json
{
"mcpServers": {
"opentargets-server": {
"command": "node",
"args": ["/path/to/opentargets-server/build/index.js"]
}
}
}
```
### Other MCP Clients
```bash
node /path/to/opentargets-server/build/index.js
```
## 🛠️ **Available Tools**
### 🎯 `search_targets`
**Search therapeutic targets by gene symbol, name, or description**
```json
{
"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**
```json
{
"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**
```json
{
"name": "get_target_disease_associations",
"arguments": {
"targetId": "ENSG00000012048", // Target Ensembl ID
"size": 10 // Optional: 1-500 results
}
}
```
**OR**
```json
{
"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**
```json
{
"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**
```json
{
"name": "get_target_details",
"arguments": {
"id": "ENSG00000012048" // Target Ensembl gene ID
}
}
```
### 🎭 `get_disease_details`
**Get comprehensive disease information**
```json
{
"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 information
- `opentargets://disease/{efoId}` - Complete disease information
- `opentargets://drug/{chemblId}` - Drug information
- `opentargets://association/{targetId}/{diseaseId}` - Association evidence
- `opentargets://search/{query}` - Search results
## 🧪 **Real-World Examples**
### Cancer Research Workflow
```bash
# 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
```bash
# 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](https://augmentednature.ai). For enhancements:
1. Fork the repository
2. Make your changes
3. Submit a pull request
## **Support**
For issues with:
- **MCP Server**: Check the server logs and error outputs
- **Open Targets API**: Visit [platform.opentargets.org](https://platform.opentargets.org/)
- **GraphQL Queries**: Use the [Open Targets GraphQL browser](https://api.platform.opentargets.org/api/v4/graphql/browser)
## Citation
If you use this project in your research or publications, please cite it as follows:
```bibtex @misc{opentargetsmcp2025,
author = {Moudather Chelbi},
title = {OpenTargets MCP Server},
year = {2025},
howpublished = {https://github.com/Augmented-Nature/OpenTargets-MCP-Server},
note = {Accessed: 2025-06-29}
