MCP Catalogs
Home

biomcp

by genomoncology·507·Score 53

BioMCP is a biomedical-focused MCP server that provides unified access to multiple biomedical data sources through a single command interface.

knowledge-graphai-llmdeveloper-tools
102
Forks
0
Open issues
1 mo ago
Last commit
2d ago
Indexed

Overview

BioMCP cuts through the biomedical data maze by providing a single command grammar to search, focus, and pivot across various biomedical sources that normally live behind different APIs and search interfaces. It supports 13 different entity types including genes, variants, articles, trials, drugs, diseases, pathways, proteins, and more. The server can be run as a local binary, through PyPI, or as an HTTP server, and integrates with Claude Desktop as an extension. BioMCP provides features like literature search across PubTator3 and Europe PMC, entity pivoting, study analysis, and enrichment capabilities.

Try asking AI

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

you:Researchers performing literature reviews and meta-analysis across multiple biomedical databases
you:Clinicians seeking information about gene variants, drug interactions, and disease associations
you:Biomedical developers building applications that need to query multiple biomedical APIs
you:What biomedical sources does BioMCP integrate with?
you:Does BioMCP require API keys to function?

When to choose this

Choose BioMCP when you need unified access to multiple biomedical databases and literature sources for research, clinical queries, or bioinformatics analysis without learning multiple APIs.

When NOT to choose this

Don't choose BioMCP if you need write access to databases (it's read-only), or if you work primarily with non-biomedical data domains outside its specialized scope.

Tools this server exposes

12 tools extracted from the README
  • searchsearch <entity> [filters]

    Discovery search across biomedical entities

  • suggestsuggest <question>

    Route biomedical questions to worked examples and starter commands

  • getget <entity> <id> [sections]

    Get detailed information about a specific biomedical entity

  • enrichenrich <GENE1,GENE2,...>

    Perform gene-set enrichment analysis using g:Profiler

  • batchbatch <entity> <id1,id2,...>

    Get details for multiple entities in parallel

  • discoverdiscover <query>

    Resolve biomedical concepts before entity selection

  • studystudy <command>

    Local study analytics for cBioPortal-style datasets

  • listlist <entity>

    List available options for an entity type

  • healthhealth

    Check API connectivity and status

  • skillskill <command>

    Manage BioMCP agent skills and workflows

  • drug interactionsdrug interactions <drug>

    Find drug interactions for a specific medication

  • article entitiesarticle entities <article_id>

    Extract entities from a specific article

Comparable tools

medprompt-mcpclinical-mcpbioinformatics-clipubmed-api

Installation

Binary install

curl -fsSL https://biomcp.org/install.sh | bash

PyPI tool install

uv tool install biomcp-cli
# or: pip install biomcp-cli

Claude Desktop extension

Install BioMCP from the Anthropic Directory in Claude Desktop.

MCP clients

{
  "mcpServers": {
    "biomcp": {
      "command": "biomcp",
      "args": ["serve"]
    }
  }
}

Remote HTTP server

biomcp serve-http --host 127.0.0.1 --port 8080

FAQ

What biomedical sources does BioMCP integrate with?
BioMCP integrates with 13+ biomedical data sources including PubTator3, Europe PMC, MyGene.info, UniProt, Reactome, ClinicalTrials.gov, and many others depending on the entity type.
Does BioMCP require API keys to function?
Most commands work without credentials, but optional API keys can improve rate limits or unlock additional features like Semantic Scholar access, OncoKB variant data, and DisGeNET associations.

Compare biomcp with

GitHub →

Last updated · Auto-generated from public README + GitHub signals.