README.md

metadata

title: HITL Drug Discovery
emoji: 🧬
colorFrom: blue
colorTo: purple
sdk: gradio
sdk_version: 5.49.1
app_file: app.py
pinned: false
hf_oauth: true
hf_oauth_scopes:
  - inference-api
license: apache-2.0
short_description: HITL Drug Discovery with AI Chat and Molecular Analysis

🧬 HITL Drug Discovery Platform

An advanced drug discovery platform combining AI-powered chat with molecular visualization and property analysis using Gradio, RDKit, and the Hugging Face Inference API.

🚀 Quick Start

Running the Application

# Easy launcher (recommended)
python run.py

# Direct launch
python app.py

# Legacy version (backup)
python legacy/app_legacy.py

Access

Once running, open your browser to: http://localhost:7860

✨ Key Features

• 🤖 AI Chat Assistant: Expert medicinal chemistry guidance and structure generation
• 🔬 Molecular Analysis: Calculate drug-likeness properties and Lipinski's Rule of Five
• 🎨 Interactive Visualization: High-quality molecular structure rendering with multiple styles
• 📚 Drug Discovery Library: Curated collection of pharmaceutical compounds
• ⚡ Real-time Property Calculation: Molecular weight, LogP, TPSA, hydrogen bonding, and more
• 🔄 Chemical Variations: Generate multiple visualization styles of molecular structures
• 💾 Bookmarking: Save and manage favorite molecular structures

🏗️ Project Structure

HITL_Drug_Discovery/
├── app.py                  # Main entry point
├── run.py                  # Easy launcher script
├── requirements.txt        # Dependencies
├── src/                   # Modular source code
│   ├── app.py             # Main application orchestrator
│   ├── molecules/         # Molecular analysis and variations
│   │   ├── analysis.py    # Property calculations & validation
│   │   └── variations.py  # Structure variation generation
│   ├── ai/                # AI services
│   │   └── services.py    # AI chat and structure generation
│   ├── ui/                # UI components and handlers
│   │   ├── components.py  # UI component definitions
│   │   └── handlers.py    # Event handlers and business logic
│   └── config/            # Configuration and settings
│       └── settings.py    # App configuration and styling
└── legacy/                # Legacy monolithic version (backup)
    └── app_legacy.py

🔧 Technical Architecture

Modular Design Benefits

• Maintainability: Each module has a single responsibility
• Testability: Individual modules can be tested in isolation
• Scalability: Easy to add new features without affecting existing code
• Readability: Clear separation of concerns
• Reusability: Components can be reused across different parts of the app

Key Modules

• molecules/analysis.py: SMILES validation, property calculations, drug-likeness assessment
• molecules/variations.py: Generation of multiple molecular structure visualizations
• ai/services.py: AI-powered chat responses using Hugging Face models
• ui/components.py: Reusable UI component definitions for Gradio interface
• ui/handlers.py: Event handlers, business logic, and state management

🛠️ Troubleshooting

Virtual Environment Issues

# Create virtual environment
python -m venv venv

# Activate it
source venv/bin/activate  # On Windows: venv\Scripts\activate

# Install dependencies
pip install -r requirements.txt

Common Issues

• Import Errors: Make sure you're in the project directory and virtual environment is activated
• Port Already in Use: The app will automatically use the next available port
• Missing Dependencies: Run pip install -r requirements.txt

🚀 Future Enhancements

With the modular structure, it's easy to:

• Add new molecular analysis algorithms
• Implement additional AI models
• Create new UI components
• Add database integration
• Implement user authentication
• Add API endpoints
• Create unit tests for each module