BioMCP

BioMCP: Biomedical Model Context Protocol

BioMCP is an open source (MIT License) toolkit that empowers AI assistants and agents with specialized biomedical knowledge. Built following the Model Context Protocol (MCP), it connects AI systems to authoritative biomedical data sources, enabling them to answer questions about clinical trials, scientific literature, and genomic variants with precision and depth.

MCPHub Certification

BioMCP is certified by MCPHub. This certification ensures that BioMCP follows best practices for Model Context Protocol implementation and provides reliable biomedical data access.

Why BioMCP?

While Large Language Models have broad general knowledge, they often lack specialized domain-specific information or access to up-to-date resources. BioMCP bridges this gap for biomedicine by:

• Providing structured access to clinical trials, biomedical literature, and genomic variants
• Enabling natural language queries to specialized databases without requiring knowledge of their specific syntax
• Supporting biomedical research workflows through a consistent interface
• Functioning as an MCP server for AI assistants and agents

Biomedical Data Sources

BioMCP integrates with multiple biomedical data sources:

Literature Sources

• PubTator3/PubMed - Peer-reviewed biomedical literature with entity annotations
• bioRxiv/medRxiv - Preprint servers for biology and health sciences
• Europe PMC - Open science platform including preprints

Clinical & Genomic Sources

• ClinicalTrials.gov - Clinical trial registry and results database
• NCI Clinical Trials Search API - National Cancer Institute's curated cancer trials database
  • Advanced search filters (biomarkers, prior therapies, brain metastases)
  • Organization and intervention databases
  • Disease vocabulary with synonyms
• BioThings Suite - Comprehensive biomedical data APIs:
  • MyVariant.info - Consolidated genetic variant annotation
  • MyGene.info - Real-time gene annotations and information
  • MyDisease.info - Disease ontology and synonym information
  • MyChem.info - Drug/chemical annotations and properties
• TCGA/GDC - The Cancer Genome Atlas for cancer variant data
• 1000 Genomes - Population frequency data via Ensembl
• cBioPortal - Cancer genomics portal with mutation occurrence data

Regulatory & Safety Sources

• OpenFDA - FDA regulatory and safety data:
  • Drug Adverse Events (FAERS) - Post-market drug safety reports
  • Drug Labels (SPL) - Official prescribing information
  • Device Events (MAUDE) - Medical device adverse events, with genomic device filtering

Available MCP Tools

BioMCP provides 24 specialized tools for biomedical research:

Core Tools (3)

1. Think Tool (ALWAYS USE FIRST!)

CRITICAL: The think tool MUST be your first step for ANY biomedical research task.

The sequential thinking tool helps:

• Break down complex biomedical problems systematically
• Plan multi-step research approaches
• Track reasoning progress
• Ensure comprehensive analysis

2. Search Tool

The search tool supports two modes:

Unified Query Language (Recommended)

Use the query parameter with structured field syntax for powerful cross-domain searches:

Supported Fields:

• Cross-domain: gene:, variant:, disease:
• Trials: trials.condition:, trials.phase:, trials.status:, trials.intervention:
• Articles: articles.author:, articles.journal:, articles.date:
• Variants: variants.significance:, variants.rsid:, variants.frequency:

Domain-Based Search

Use the domain parameter with specific filters:

Note: When searching articles with a gene parameter, cBioPortal data is automatically included:

• Gene-level summaries show mutation frequency across cancer studies
• Mutation-specific searches (e.g., "V600E") show study-level occurrence data
• Cancer types are dynamically resolved from cBioPortal API

3. Fetch Tool

Retrieve full details for a single article, trial, or variant:

Domain-specific options:

• Articles: detail="full" retrieves full text if available
• Trials: detail can be "protocol", "locations", "outcomes", "references", or "all"
• Variants: Always returns full details

Individual Tools (21)

For users who prefer direct access to specific functionality, BioMCP also provides 21 individual tools:

Article Tools (2)

• article_searcher: Search PubMed/PubTator3 and preprints
• article_getter: Fetch detailed article information (supports PMID and DOI)

Trial Tools (5)

• trial_searcher: Search ClinicalTrials.gov or NCI CTS API (via source parameter)
• trial_getter: Fetch all trial details from either source
• trial_protocol_getter: Fetch protocol information only (ClinicalTrials.gov)
• trial_references_getter: Fetch trial publications (ClinicalTrials.gov)
• trial_outcomes_getter: Fetch outcome measures and results (ClinicalTrials.gov)
• trial_locations_getter: Fetch site locations and contacts (ClinicalTrials.gov)

Variant Tools (2)

• variant_searcher: Search MyVariant.info database
• variant_getter: Fetch comprehensive variant details

NCI-Specific Tools (6)

• nci_organization_searcher: Search NCI's organization database
• nci_organization_getter: Get organization details by ID
• nci_intervention_searcher: Search NCI's intervention database (drugs, devices, procedures)
• nci_intervention_getter: Get intervention details by ID
• nci_biomarker_searcher: Search biomarkers used in trial eligibility criteria
• nci_disease_searcher: Search NCI's controlled vocabulary of cancer conditions

Gene, Disease & Drug Tools (3)

• gene_getter: Get real-time gene information from MyGene.info
• disease_getter: Get disease definitions and synonyms from MyDisease.info
• drug_getter: Get drug/chemical information from MyChem.info

Note: All individual tools that search by gene automatically include cBioPortal summaries when the include_cbioportal parameter is True (default). Trial searches can expand disease conditions with synonyms when expand_synonyms is True (default).

Quick Start

For Claude Desktop Users

1. Install uv if you don't have it (recommended):
   # MacOS brew install uv # Windows/Linux pip install uv
2. Configure Claude Desktop:
   • Open Claude Desktop settings
   • Navigate to Developer section
   • Click "Edit Config" and add:
   { "mcpServers": { "biomcp": { "command": "uv", "args": ["run", "--with", "biomcp-python", "biomcp", "run"] } } }
   • Restart Claude Desktop and start chatting about biomedical topics!

Python Package Installation

Configuration

Environment Variables

BioMCP supports optional environment variables for enhanced functionality:

Running BioMCP Server

BioMCP supports multiple transport protocols to suit different deployment scenarios:

Local Development (STDIO)

For direct integration with Claude Desktop or local MCP clients:

HTTP Server Mode

BioMCP supports multiple HTTP transport protocols:

Legacy SSE Transport (Worker Mode)

For backward compatibility with existing SSE clients:

Streamable HTTP Transport (Recommended)

The new MCP-compliant Streamable HTTP transport provides optimal performance and standards compliance:

Features of Streamable HTTP transport:

• Single /mcp endpoint for all operations
• Dynamic response mode (JSON for quick operations, SSE for long-running)
• Session management support (future)
• Full MCP specification compliance (2025-03-26)
• Better scalability for cloud deployments

Deployment Options

Docker

Cloudflare Workers

The worker mode can be deployed to Cloudflare Workers for global edge deployment.

Note: All APIs work without authentication, but tokens may provide higher rate limits.

Command Line Interface

BioMCP provides a comprehensive CLI for direct database interaction:

Testing & Verification

Test your BioMCP setup with the MCP Inspector:

This opens a web interface where you can explore and test all available tools.

Enterprise Version: OncoMCP

OncoMCP extends BioMCP with GenomOncology's enterprise-grade precision oncology platform (POP), providing:

• HIPAA-Compliant Deployment: Secure on-premise options
• Real-Time Trial Matching: Up-to-date status and arm-level matching
• Healthcare Integration: Seamless EHR and data warehouse connectivity
• Curated Knowledge Base: 15,000+ trials and FDA approvals
• Sophisticated Patient Matching: Using integrated clinical and molecular profiles
• Advanced NLP: Structured extraction from unstructured text
• Comprehensive Biomarker Processing: Mutation and rule processing

Learn more: GenomOncology

MCP Registries

Example Use Cases

Gene Information Retrieval

Disease Synonym Expansion

Integrated Biomedical Research

Documentation

For comprehensive documentation, visit https://biomcp.org

Developer Guides

• HTTP Client Guide - Using the centralized HTTP client
• Migration Examples - Migrating from direct HTTP usage
• Error Handling Guide - Comprehensive error handling patterns
• Integration Testing Guide - Best practices for reliable integration tests
• Third-Party Endpoints - Complete list of external APIs used
• Testing Guide - Running tests and understanding test categories

Development

Running Tests

Note: Integration tests make real API calls and may fail due to network issues or rate limiting. In CI/CD, integration tests are run separately and allowed to fail without blocking the build.

BioMCP Examples Repo

Looking to see BioMCP in action?

Check out the companion repository: 👉 biomcp-examples

It contains real prompts, AI-generated research briefs, and evaluation runs across different models. Use it to explore capabilities, compare outputs, or benchmark your own setup.

Have a cool example of your own? We’d love for you to contribute! Just fork the repo and submit a PR with your experiment.

License

This project is licensed under the MIT License.