Enabling AI to speak the language of biology.
BioContext is an open-source collection of Model Context Protocol (MCP) servers that wrap major bio-databases and make them instantly query-able by LLMs and agent frameworks.
| Status | Repo | Data Source | Quick blurb | Last update* |
|---|---|---|---|---|
| ✅ | BioMart-MCP | BioMart | Query federated genomic datasets via BioMart. | 2025-05-03 |
| ✅ | PubMed-MCP | PubMed | Search & retrieve biomedical citations. | 2025-05-01 |
| ✅ | OpenTargets-MCP | Open Targets Platform | Drug-target evidence & associations. | 2025-04-21 |
| ✅ | AACT-MCP | AACT / ClinicalTrials.gov | Access aggregated clinical-trials data. | 2025-04-21 |
| ✅ | AlphaFold-MCP | AlphaFold DB | Retrieve predicted protein structures. | 2025-04-18 |
| ✅ | UniProt-MCP | UniProt | Protein sequences & annotations. | 2025-04-17 |
| ✅ | ChemBL-MCP | ChEMBL | Bioactivity / drug-like molecules. | 2025-04-17 |
| ✅ | PubChem-MCP | PubChem | Compounds, substances & assays. | 2025-04-17 |
*“Last update” = last push on GitHub at time of this README refresh (7 May 2025).
Looking for install instructions?
Every repo ships its own README withpip/uv/Docker commands and examples.
MCP is an open specification that lets external tools expose structured resources, tools, and prompts to AI assistants over simple transports (STDIO, HTTP, SSE). Wrapping bio-databases behind MCP servers means:
- Zero-boilerplate integration with clients such as Claude Desktop, Cursor IDE, or AutoGen agents.
- Schema discovery – each tool is self-describing via JSON Schema.
- Language-agnostic – works with Python, Node, Go, Rust, … anything that can read/write JSON.
# Example: spin up PubChem-MCP in a fresh venv
uv venv && source .venv/bin/activate
pip install pubchem-mcp
pubchem-mcp --stdio # now point your MCP-aware client at STDIO
