NCBI-Database-MCP

NCBI Database MCP

🔍 MCP server for NCBI bioinformatics tools and disease-focused gene expression research

Enable AI assistants to discover gene expression datasets by disease/condition and access comprehensive NCBI databases through natural language. Perfect for researchers studying disease mechanisms and therapeutic targets.

🧬 Features

• 🔬 Disease-Focused GEO Search - Discover gene expression datasets by disease/condition and organism
• 📊 Comprehensive Study Metadata - Get detailed methodology, platform, and sample information
• 🧬 Gene-to-Genomic Conversion - Convert gene names to genomic DNA sequences
• 🐭 Multi-Species Support - Human, mouse, and rat datasets
• 📈 Research Methodology Details - RNA-Seq, microarray, ChIP-Seq, and other techniques
• 🔗 Direct Database Links - Easy access to full datasets and original studies

🚀 Quick Start

Installation

# Clone repository
git clone https://github.com/hpend2373/NCBI-Database-MCP.git
cd NCBI-Database-MCP

# Install dependencies
pip install -r requirements.txt

Basic Usage

🚀 RECOMMENDED: Use FastMCP Server for Best Performance

# Start the FastMCP server (RECOMMENDED)
./run_fastmcp_gene_server.sh

# Alternative: Standard MCP server (slower startup)
python src/gene_to_genomic_server.py

Why FastMCP?

• ⚡ Faster startup - Instant server initialization
• 🔧 Easier debugging - Better error messages and logging
• 📊 Built-in monitoring - Performance metrics included
• 🎯 Optimized for research - Designed specifically for bioinformatics workflows

Configuration

Add to your MCP client config:

{
  "mcpServers": {
    "ncbi-database": {
      "command": "python",
      "args": ["src/gene_to_genomic_server.py"],
      "cwd": "/path/to/NCBI-Database-MCP",
      "env": {
        "NCBI_API_KEY": "your_api_key_here"
      }
    }
  }
}

Alternative: Set global environment variable

export NCBI_API_KEY="your_api_key_here"

Then use simpler config:

{
  "mcpServers": {
    "ncbi-database": {
      "command": "python",
      "args": ["src/gene_to_genomic_server.py"],
      "cwd": "/path/to/NCBI-Database-MCP"
    }
  }
}

💡 Usage Examples

🔬 Disease Expression Research (Primary Use Case)

User: "Find gene expression datasets for Alzheimer's disease in humans"
AI: [calls search_geo_datasets] → 
📊 Returns 10 datasets with:
- Study methodology (RNA-Seq, Microarray)
- Sample sizes and experimental design
- Platform information (Illumina, Affymetrix)
- Research summaries and direct GEO links

User: "Show me cancer expression studies in mice using RNA sequencing"
AI: [calls search_geo_datasets] → 
🧪 Filtered results showing:
- RNA-Seq datasets only
- Mouse-specific cancer studies
- Detailed experimental protocols

🧬 Gene-to-Genomic Analysis

User: "Get the genomic sequence for BRCA1"
AI: [calls gene_to_genomic_sequence] → Returns genomic DNA sequence in FASTA format

📍 Gene Information & Location

User: "Find information about TP53 gene"
AI: [calls search_gene_info] → Returns gene location, function, and coordinates

🎯 Coordinate-Based Sequence Retrieval

User: "Get sequence from chr17:43044295-43125483"
AI: [calls get_genomic_sequence] → Returns DNA sequence for specified coordinates

🛠️ Available Tools

🔬 search_geo_datasets (Primary Tool)

Discover gene expression datasets by disease/condition and organism

Parameters:

• disease (required) - Disease or condition name
  • Examples: "cancer", "diabetes", "Alzheimer", "heart disease", "depression"
• organism - Target organism (default: "Homo sapiens")
  • Options: "Homo sapiens", "Mus musculus", "Rattus norvegicus"
• study_type - Expression study methodology (optional, default: "Expression profiling by high throughput sequencing")
  • Options: "Expression profiling by array", "Expression profiling by high throughput sequencing"
  • Default: RNA-Seq - Most comprehensive and current sequencing technology
• max_results - Maximum results to return (1-50, default: 10)

Detailed Output:

• 📊 Dataset Information: GDS accession numbers and titles
• 🔬 Study Methodology:
  • RNA-Seq (High-throughput transcriptome sequencing) - DEFAULT
  • Microarray (Hybridization-based gene expression)
  • ChIP-Seq (Chromatin immunoprecipitation sequencing)
  • SAGE (Serial analysis of gene expression)
• 🧬 Data Type Classification:
  • Single-Cell RNA-Seq 🧩 - Individual cell-level gene expression
  • Bulk RNA-Seq 📦 - Tissue/population-level gene expression
  • Spatial Transcriptomics 🗺️ - Location-aware gene expression
• 🧪 Platform Details: Illumina, Affymetrix, Agilent technologies
• 📈 Experimental Design: Sample counts, tissue types, treatment conditions
• 📝 Research Context: Study summaries and disease relevance
• 🔗 Direct Access: Links to full datasets on NCBI GEO

🧬 gene_to_genomic_sequence

Convert gene name to genomic DNA sequence

Parameters:

• gene_name (required) - Gene symbol (e.g., "BRCA1", "TP53")
• organism - Target organism (default: "human")
• sequence_type - "genomic", "cds", "mrna", "protein"
• output_format - "fasta", "genbank", "json"

📍 search_gene_info

Search for gene information and genomic location

Parameters:

• gene_name (required) - Gene symbol or name
• organism - Target organism (default: "human")

🎯 get_genomic_sequence

Get genomic sequence from chromosome coordinates

Parameters:

• chromosome (required) - Chromosome accession (e.g., "NC_000017.11")
• start (required) - Start position
• end (required) - End position
• output_format - "fasta", "json"

⚙️ Configuration

Environment Variables

You can configure the server using environment variables:

# Copy example file and edit
cp .env.example .env

# Or set directly
export NCBI_API_KEY="your_api_key_here"

# Get your free API key from: https://www.ncbi.nlm.nih.gov/account/
# Without API key: 3 requests/second
# With API key: 10 requests/second

📁 Project Structure

NCBI-Database-MCP/
├── README.md                    # Documentation
├── requirements.txt             # Python dependencies
├── pyproject.toml              # Project configuration
├── .env.example                # Environment variables template
├── run_fastmcp_gene_server.sh  # Launch script
└── src/
    ├── gene_to_genomic_server.py  # Standard MCP server
    └── fastmcp_gene_server.py     # FastMCP server (recommended)

📈 Performance Tips

🔬 GEO Dataset Search Optimization

• Use specific disease terms: "lung cancer" > "cancer", "type 2 diabetes" > "diabetes"
• Combine with study types: Filter by methodology for targeted results
• Start with small result sets: Use max_results=5-10 for initial exploration
• Organism specificity: Use exact names ("Homo sapiens" not "human")

🐛 Troubleshooting

Common Issues

Gene not found

# Check gene name spelling
# Try alternative gene symbols
# Verify organism specification

No GEO datasets found

# Try broader disease terms (e.g., "cancer" instead of "lung adenocarcinoma")
# Check organism name (use "Homo sapiens" not "human")
# Try without study_type filter
# Verify disease spelling and terminology

API rate limiting

# Get free NCBI API key: https://www.ncbi.nlm.nih.gov/account/
# Set NCBI_API_KEY environment variable
# Without key: 3 requests/second limit
# With key: 10 requests/second limit

Network timeouts

# Check internet connection
# Increase timeout values
# Retry failed requests

📚 Resources

• Gene Database
• BLAST Documentation

🆘 Support

• 🐛 Bug Reports: GitHub Issues
• 💡 Feature Requests: GitHub Issues
• 📖 Documentation: README.md

---

Happy genomics research! 🧬🔍