🧬 BioBTree v2

A unified biomedical graph database that integrates 50+ primary data sources — genes, proteins, compounds, diseases, pathways, variants, expression, and more — into a single queryable graph with billions of cross-reference edges, accessible by both researchers and AI.

BRCA1 >> ensembl >> uniprot >> pdb[resolution<2.0]

This finds BRCA1 in Ensembl, maps to UniProt proteins, and returns high-resolution PDB structures — crossing three databases in a single line.

🔗 Try It

The fastest way to experience BioBTree v2 is through an AI assistant with MCP (Model Context Protocol). We recommend Claude CLI (tested extensively), though Codex CLI and Gemini CLI also work:

{
  "mcpServers": {
    "biobtree": {
      "type": "http",
      "url": "https://sugi.bio/biobtree/mcp"
    }
  }
}

Once connected, just ask questions in natural language — the AI will query BioBTree automatically:

💊 "What tissues express SCN9A most highly? Are there safety concerns for a Nav1.7 inhibitor?"

🧪 "How many ClinVar variants does BRCA1 have? How many are pathogenic?"

🎯 "What are all the protein targets of Alectinib with IC50 values?"

A REST API is also available for direct programmatic access:

https://sugi.bio/biobtree/api/ws/?i=BRCA1
https://sugi.bio/biobtree/api/ws/map/?i=BRCA1&m=>>ensembl>>uniprot>>chembl_target
https://sugi.bio/biobtree/api/ws/entry/?i=P38398&s=uniprot

📖 Documentation

Query syntax, integrated databases, development, and other details check the : docs/ or refer to latest preprint.

For any questions, issues, or collaboration ideas, feel free to create an issue or reach out at tamer.gur07@gmail.com.

📄 Publication

BioBTree v2: Grounding LLM Responses with Large-Scale Structured Biomedical Data Preprint: [link forthcoming] BioBTree v1: F1000Research

⚖️ License

GPL-v3