Skip to content
Conduit

mcp server

Mini-PharmaMind

An intelligent multi-agent system for pharmaceutical research and drug discovery, powered by AI. Mini-PharmaMind orchestrates specialized agents to analyze targets, search for drugs, and generate comp

khireddinemahaline/Mini-PharmaMindcommunitylistedunreviewed
Use tool

Set up

Runs on your computer. Pick your agent and operating system.

This project does not publish an install method we can turn into a config. Follow the README instead.

README

synced Oct 22, 2026

Mini-PharmaMind 🧬💊

⚠️ Mini Version Notice: This is a proof-of-concept limited version of PharmaMind for educational and research purposes.

An intelligent multi-agent system for pharmaceutical research and drug discovery, powered by AI. Mini-PharmaMind orchestrates specialized agents to analyze targets, search for drugs, and generate comprehensive research reports.

License: MIT Python 3.10+ Version

🌟 Features

  • Multi-Agent Orchestration: Specialized agents working together for drug discovery workflows
  • Target Search Agent: Analyzes biological targets using ChEMBL and OpenTargets databases
  • Drug Search Agent: Discovers and evaluates potential drug candidates
  • Report Generation: Creates comprehensive PDF reports with findings
  • Interactive Chat Interface: User-friendly Chainlit interface for conversational AI
  • Real-time Streaming: Stream responses and updates in real-time
  • Session Management: Persistent state across conversations
  • Arize Phoenix Observability: Full multi-agent tracing with OpenTelemetry and OpenInference instrumentation

⚠️ Current Limitations (Mini Version)

This mini version has the following limitations:

Architecture

  • ✅ Functional Tools Implementation - Uses direct API calls instead of MCP servers for simplicity
  • ⚠️ Limited Test Coverage - Test functions are basic and need expansion
  • ⚠️ Performance Issues - Higher latency, optimization needed
  • ⚠️ Basic Tool Set - Limited tools compared to full version

Missing Features (Available in Full Version)

  • ❌ Advanced RAG System - Needed for real pharmaceutical laboratory analysis
  • ❌ Drug-Target Simulation - Molecular dynamics and docking simulations
  • ❌ Physicochemical Prediction - AI-powered ADMET and property predictions
  • ❌ MCP Server Architecture - Scalable Model Context Protocol integration
  • ❌ Real-time Lab Integration - Connection with laboratory equipment
  • ❌ Advanced Caching - Performance optimization layer

🚀 Future Enhancements (Full Version Roadmap)

We welcome collaboration on developing the full PharmaMind version with:

🧪 Advanced Computational Chemistry

  • Molecular Dynamics Simulations - Drug-target interaction modeling
  • Binding Affinity Prediction - AI-powered binding score calculations
  • ADMET Property Prediction - Absorption, Distribution, Metabolism, Excretion, Toxicity
  • Physicochemical Parameter Prediction - LogP, solubility, permeability, etc.
  • Molecular Docking - Automated docking with multiple scoring functions

🤖 Enhanced AI Capabilities

  • RAG System - Retrieval-Augmented Generation for pharmaceutical literature
  • Multi-Modal Analysis - Process scientific papers, patents, and molecular structures
  • Knowledge Graph Integration - Drug-disease-target relationship mapping
  • Predictive Models - Machine learning for drug efficacy and safety

⚡ Performance & Scalability

  • MCP Server Architecture - Distributed agent system
  • Response Caching - Reduce API calls and improve latency
  • Parallel Processing - Concurrent agent execution
  • Optimized Database - Advanced querying and indexing

🔬 Laboratory Integration

  • LIMS Integration - Laboratory Information Management Systems
  • Instrument Data - Real-time data from analytical instruments
  • Electronic Lab Notebooks - Seamless documentation
  • Workflow Automation - End-to-end laboratory process automation

🤝 Collaboration Opportunities

We're looking for collaborators in:

  • Computational Chemistry - Molecular simulation experts
  • Machine Learning - ML model development for drug discovery
  • Pharmaceutical Research - Domain experts for validation
  • Software Engineering - Scalability and performance optimization
  • Data Science - Large-scale data processing and analysis

Interested in contributing? See CONTRIBUTING.md for guidelines.

📋 Prerequisites

  • Python 3.10 or higher
  • uv package manager (recommended) or pip
  • Docker and Docker Compose (for containerized deployment)
  • PostgreSQL (for Prisma database)

🚀 Quick Start

Installation

  1. Clone the repository

    git clone https://github.com/khireddinemahaline/Mini-PharmaMind.git
cd Mini-PharmaMind/pharma-core

  2. Run the setup script (recommended)

    chmod +x setup.sh
./setup.sh

    Or manually:

  3. Set up environment variables

    cp .env.example .env
# Edit .env with your API keys and configuration

    Required environment variables:

    # LLM Provider (openai or deepseek)
LLM_PROVIDER=openai
OPENAI_API_KEY=your_openai_api_key_here

# Database
DATABASE_URL=postgresql://user:password@localhost:5432/pharmadb

# Chainlit Configuration
CHAINLIT_AUTH_SECRET=your_secret_key_here

# Arize Phoenix Observability
ARIZE_SPACE_ID=your_arize_space_id
ARIZE_API_KEY=your_arize_api_key

  4. Install dependencies

    uv sync

    Or with pip:

    pip install -e .

  5. Set up the database

    npx prisma generate
npx prisma db push

  6. Run the application

    chainlit run orcastration/main_chainlit.py -w --host 0.0.0.0 --port 8000

    Access the interface at http://localhost:8000

🔐 Default Test Credentials

For testing purposes, you can use these default credentials:

  • Username: researcher
  • Password: easydiscovery##1

⚠️ Security Note: These are default credentials for testing only. In production, change these immediately and use strong passwords.

🐳 Docker Deployment

Quick Start with Docker Compose

docker compose up --build

Development Mode

docker compose -f docker-compose_dev.yaml up --build

📁 Project Structure

pharma-core/
├── agents/                 # Agent implementations
│   ├── target_search.py   # Target search agent
│   ├── drug_search.py     # Drug search agent
│   ├── report.py          # Report generation agent
│   └── critique.py        # Critique agent
├── config/                # Configuration files
│   ├── llm_client.py      # LLM client setup
│   └── sytem_prompts.py   # System prompts
├── orcastration/          # Main orchestration
│   └── main_chainlit.py   # Chainlit interface
├── tools/                 # Agent tools and utilities
├── utilities/             # Helper functions
├── tests/                 # Test suite
├── prisma/                # Database schema
│   └── schema.prisma
├── pyproject.toml         # Project configuration
├── Dockerfile             # Docker configuration
├── setup.sh               # Automated setup script
└── README.md              # This file

🔧 Configuration

LLM Models

Configure your preferred models in .env:

# Use OpenAI (default)
LLM_PROVIDER=openai
OPENAI_API_KEY=your_key_here
LLM_MODEL=gpt-4-turbo-preview

# Or use DeepSeek
LLM_PROVIDER=deepseek
DEEPSEEK_API=your_key_here
LLM_MODEL=deepseek-chat

Agent Behavior

Customize agent prompts and behavior in config/sytem_prompts.py.

🤖 Agents

Target Search Agent

Analyzes biological targets, diseases, and pathways using:

  • ChEMBL database queries
  • OpenTargets platform integration
  • Literature search capabilities

Drug Search Agent

Discovers drug candidates through:

  • Chemical structure search
  • Similarity analysis
  • Drug-target interaction queries
  • ADMET property evaluation

Report Generation Agent

Creates comprehensive reports including:

  • Executive summaries
  • Detailed findings
  • Visualizations
  • References and citations

Critique Agent

Reviews and improves outputs by:

  • Validating scientific accuracy
  • Checking completeness
  • Suggesting improvements

🧪 Testing

Run the test suite:

pytest tests/

Run with coverage:

pytest --cov=. --cov-report=html tests/

🔐 Authentication

PharmaMind uses Chainlit's built-in authentication. Configure authentication settings in your Chainlit configuration.

📝 Development

Install development dependencies

uv sync --dev

🤝 Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

  1. Fork the repository
  2. Create your feature branch (git checkout -b feature/AmazingFeature)
  3. Commit your changes (git commit -m 'Add some AmazingFeature')
  4. Push to the branch (git push origin feature/AmazingFeature)
  5. Open a Pull Request

See CONTRIBUTING.md for detailed guidelines.

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

📚 Documentation

🙏 Acknowledgments

  • ChEMBL - Chemical database of bioactive molecules
  • OpenTargets - Target validation platform
  • AutoGen - Multi-agent conversation framework
  • Chainlit - Conversational AI interface
  • Arize Phoenix - Multi-agent observability and tracing

📧 Contact

MHLAINE Khireddine - mhalaine.khireddine.chimie@gmail.com

Project Link: https://github.com/khireddinemahaline/Mini-PharmaMind

Full Version Development: We're actively working on the full PharmaMind version with advanced features. Interested in early access or collaboration? Reach out!

🗺️ Roadmap

Near-term (Mini Version Improvements)

  • Improve test coverage (increase from basic to comprehensive)
  • Optimize latency and performance
  • Add response caching layer
  • Expand tool library
  • Add more databases (PubChem, DrugBank)
  • Enhance error handling and logging

Mid-term (Transition to Full Version)

  • Implement RAG system for pharmaceutical literature
  • Add basic physicochemical property prediction
  • Integrate MCP server architecture
  • Implement advanced visualization tools
  • Add collaborative features
  • Create REST API for programmatic access

Long-term (Full PharmaMind)

  • Molecular dynamics simulation engine
  • Drug-target docking and binding affinity prediction
  • Complete ADMET prediction suite
  • Laboratory equipment integration
  • Multi-modal AI (text, structure, images)
  • Real-time collaboration platform
  • Enterprise-grade security and compliance

Want to help? Join us in building the future of AI-powered drug discovery! 🚀

⚠️ Disclaimer

This software is for research purposes only. Always consult with qualified pharmaceutical professionals for drug discovery and development decisions.

Made with ❤️ for the pharmaceutical research community

↑↓ navigate openesc close