A beautiful, fully-functional AI chemistry assistant for material science and chemistry research.
- 🔬 Research Mode: Ask chemistry questions and get detailed explanations
- 📋 Protocol Generation: Generate step-by-step lab protocols
- 🤖 Lab Automation: Create Python scripts for Opentrons, PyHamilton
- 🛡️ Safety Analysis: Get safety information and hazard warnings
- 📊 Data Visualization: Advanced charts and graphs with offline capability
- 🧮 Calculators: Molarity, pH, reaction yield calculators
- 📝 Notes: Take and save research notes
- 📤 Export: Export protocols and data in multiple formats
- 🌙 Dark Mode: Beautiful dark/light theme toggle
- 💜 Purple Gradient Theme: Scientist-friendly UI design
./start_catalyze.sh# Activate virtual environment
source venv/bin/activate
# Start the app
python app/flask_app.pyThen open: http://localhost:5003
# Initialize submodules (first time only)
./manage_submodules.sh init
# Install Node.js dependencies
./manage_submodules.sh install
# Start the app
source venv/bin/activate && python app/flask_app.pymit-catalyze/
├── app/
│ └── flask_app.py # Main Flask backend
├── react-build/
│ └── index.html # Beautiful frontend (HTML/CSS/JS)
├── mcp_servers/
│ └── chembl-mcp-server/ # ChEMBL MCP Server (Git submodule)
├── src/
│ ├── clients/ # API clients (LLM, PubChem)
│ ├── generators/ # Protocol & automation generators
│ ├── config/ # Configuration files
│ └── pipeline.py # Main processing pipeline
├── venv/ # Python virtual environment
├── pyproject.toml # Python dependencies
├── start_catalyze.sh # Simple startup script
├── manage_submodules.sh # Submodule management script
├── .gitmodules # Git submodule configuration
└── README.md # This file
- Single HTML File: Everything runs from one beautiful
react-build/index.html - No React Build Process: Pure HTML/CSS/JavaScript - fast and simple
- Fully Functional: All features work perfectly without internet
- Beautiful UI: Purple gradient theme with smooth animations
- Clean Codebase: Minimal, focused, and maintainable
- 🧪 ChEMBL Integration: Access to 27 specialized chemistry tools via MCP
- 🤖 Hybrid AI: Combines OpenAI knowledge with ChEMBL database accuracy
This project integrates with the ChEMBL MCP Server to provide access to 27 specialized chemistry tools:
- Compound Search: Search ChEMBL database by name, synonym, or identifier
- Target Analysis: Search biological targets and get detailed information
- Bioactivity Data: Access bioactivity measurements and assay results
- Drug Development: Search approved drugs and clinical candidates
- Chemical Properties: Analyze ADMET properties and drug-likeness
- Smart Routing: Chemistry questions automatically trigger ChEMBL integration
- Hybrid Approach: OpenAI provides main answers, ChEMBL enhances with database data
- Graceful Fallback: If ChEMBL fails, you still get OpenAI responses
- Detailed Logging: Track system behavior with comprehensive logs
# Check submodule status
./manage_submodules.sh status
# Update to latest version
./manage_submodules.sh update
# Reinstall dependencies
./manage_submodules.sh install- Python: All dependencies in
pyproject.tomland installed invenv/ - Node.js: Required for ChEMBL MCP Server (installed via submodule)
- Local: http://localhost:5003
- Network: http://[your-ip]:5003
- Chat Interface: Ask chemistry questions with Research/Protocol/Automate/Safety modes
- Chat History: Create named chats, manage conversation history
- Capabilities Display: See what Catalyze can do
- Results Tabs: Calculator, Notes, Data Visualizer, Export
- Dark Mode: Toggle between light and dark themes
- Responsive Design: Works on desktop and mobile
Built for MIT Hackathon 2025 🎓