genefoundry
A modular registry for trustworthy, evidence-based genomic analysis powered by Model Context Protocols (MCPs). Bridging AI and clinical genomics with traceable, validated data retrieval.
claude mcp add --transport stdio berntpopp-genefoundry node server.js \ --env PORT="3000" \ --env NODE_ENV="production"
How to use
GeneFoundry is a modular registry designed to provide trustworthy, evidence-based genomic analysis powered by Model Context Protocols (MCPs). The landing page frontend acts as the conceptual entry point for the GeneFoundry Proof of Concept, demonstrating the methodology, data grounding approach, and user workflows. Once running, users can explore the landing page to learn how MCPs can be used to retrieve grounded genomic data, understand the evidence sources, and see how results are anchored to validated datasets. The frontend is designed to present an overview of the MCP-driven architecture, showcase example workflows, and guide users toward connected MCP-enabled tools when available. Tools and capabilities highlighted on the page focus on evidence-based retrieval, provenance tracking, and modular integration with MCP-backed data providers.
To use the site locally, start the frontend server and open the provided URL in a browser. Interact with the sections that describe the MCP-based data sources, governance, and evidence links. If you have MCP-enabled endpoints configured, you can navigate to any integrated data provider, query capabilities, or proof-of-concept demonstrations that showcase how grounding reduces hallucination and improves trust in genomic analysis outputs.
How to install
Prerequisites:
- Node.js (≥14.x) and npm or yarn installed
- Git
Installation steps:
-
Clone the repository: git clone https://github.com/berntpopp/genefoundry.git cd genefoundry
-
Install dependencies: npm install
or if you prefer yarn:
yarn install
-
Build or prepare the frontend (if applicable): npm run build
or follow project-specific build steps if defined in package.json
-
Run the frontend server: npm start
This will start the Node.js server (as configured in server.js) on port 3000 by default
-
Open http://localhost:3000 in your browser to view the GeneFoundry landing page.
Notes:
- If your environment uses a different port or host, adjust the PORT and HOST settings accordingly.
- Ensure any required environment variables for MCP integration (e.g., data source endpoints, API keys) are provided in a .env file or in your hosting environment.
Additional notes
Tips and common considerations:
- Since this is a frontend landing page for GeneFoundry, the MCP integration points (data providers, evidence sources) should be wired up in the backend or via environment-configured endpoints. Ensure the MCP endpoints are accessible and have proper CORS settings if the frontend makes cross-origin requests.
- If you encounter issues with loading data or MCP-powered components, check the browser console and server logs for errors related to data sources, authentication, or provenance tracking.
- Document any environment variables used for MCP integration (e.g., MCP_ENDPOINT, API_KEYS, DATA_SOURCE_URL) so you can reproduce the environment in staging or production.
- For production deployment, consider bundling and serving static assets behind a web server or CDN, with environment-specific configurations for MCP endpoints.
- The live site at https://genefoundry.org/ provides a demonstration of the concept; when running locally, the same MCP-driven grounding principles apply to the local frontend.
Related MCP Servers
biomcp
BioMCP: Biomedical Model Context Protocol
CanvasMCPClient
Canvas MCP Client is an open-source, self-hostable dashboard application built around an infinite, zoomable, and pannable canvas. It provides a unified interface for interacting with multiple MCP (Model Context Protocol) servers through a flexible, widget-based system.
opentargets
MCP server for Open Targets Data
ensembl
A full-featured Model Context Protocol (MCP) server that exposes Ensembl’s REST API. Built using the TypeScript MCP SDK
mcp-turso
MCP server for interacting with Turso-hosted LibSQL databases
agentgate
Approval workflows for AI agents