Bio-Agents MCP
@dogeplusplus
MCP servers for Protein Data Bank, ChemBL, and other life science data (WIP), with Ollama client for local testing.
Overview
What is Bio-Agents MCP?
Bio-Agents MCP is a collection of microservices and clients for natural language interaction with biological databases, including the Protein Data Bank (PDB) and ChEMBL chemical database, built using the FastMCP framework.
How to use Bio-Agents MCP?
Copy the example environment file, build Docker images with make build, start all services with make up, then launch the web interface with make run-chainlit and visit http://localhost:8000 to query biological data.
Key features of Bio-Agents MCP
- LLM client with web UI (Chainlit) and terminal modes
- PDB MCP Server for Protein Data Bank API
- ChEMBL MCP Server for chemical database API
- Docker Compose to run all services together
- Makefile for common tasks (build, up, down, restart)
- Each module can run independently or together
Use cases of Bio-Agents MCP
- Query protein structures and annotations from PDB using natural language
- Search chemical compounds and bioactivity data via ChEMBL
- Combine biological and chemical data through an LLM interface
FAQ from Bio-Agents MCP
What components does it include?
It includes an LLM Client (web UI and terminal), PDB MCP Server, and ChEMBL MCP Server.
How do I start the system?
Configure environment with cp .env.example .env, then run make build and make up, followed by make run-chainlit.
What transport does it use?
The architecture uses FastMCP framework; services communicate via Docker networking (implied by Docker Compose).
What are the runtime requirements?
Docker and Docker Compose are required to run all services together. Each service has its own Dockerfile.
Where are configuration files located?
Configuration files are in the conf/ directory. Detailed documentation for each service is in its own README.
