OpenPeerAI/DockingAtHOME

Docking@HOME

Distributed and Parallel Molecular Docking Platform

Overview

Docking@HOME is a cutting-edge distributed computing platform that leverages the power of volunteer computing, GPU acceleration, decentralized networking, and AI-driven orchestration to perform large-scale molecular docking simulations. This project combines multiple state-of-the-art technologies to democratize drug discovery and computational chemistry.

Key Features

• 🧬 AutoDock Integration: Uses AutoDock Suite 4.2.6 for molecular docking simulations
• 🚀 GPU Acceleration: CUDPP-powered parallel processing for enhanced performance
• 🌐 Distributed Computing: BOINC framework for volunteer computing at scale
• 🔗 Decentralized Networking: Distributed Network Settings-based coordination using the Decentralized Internet SDK
• 🤖 AI Orchestration: Cloud Agents for intelligent task distribution and optimization
• 📊 HuggingFace Integration: Model cards and datasets for reproducible research

Architecture

           Docking@HOME Platform

           ┌─────────────────────┐
           │   BOINC Server    │
           │   (Task Mgmt)     │
           └─────────▲───────────┘
                      │
                      │
           ┌──────────┴───────────┐
           │  Decentralized     │
           │     Internet       │
           └─────────▲───────────┘
                      │
                      │
           ┌──────────┴───────────┐
           │   Cloud Agents     │
           │   (AI Routing)     │
           └──────────┬───────────┘
                      │
                      ▼
    ┌──────────────────────────────────────────────┐
    │      Distributed Worker Nodes (Clients)  │
    │ ┌────────────────┐      ┌────────────────┐  │
    │ │   AutoDock    │<---->│    CUDPP     │  │
    │ │  (Docking)    │      │ (GPU Accel)  │  │
    │ └────────────────┘      └────────────────┘  │
    └──────────────────────────────────────────────┘

Components

1. AutoDock Suite (v4.2.6)

Core molecular docking engine that predicts binding modes and affinities of small molecules to protein targets.

2. CUDPP (CUDA Data Parallel Primitives Library)

Provides GPU-accelerated parallel primitives for enhancing AutoDock's computational performance.

3. BOINC (Berkeley Open Infrastructure for Network Computing)

Distributed computing middleware that manages volunteer computing resources globally.

4. The Decentralized Internet SDK

Enables Distributed Network Settings-based coordination, ensuring transparency and decentralization of task distribution.

5. Cloud Agents

AI-powered orchestration layer that optimizes task scheduling and resource allocation based on workload characteristics.

Authors & Contributors

• OpenPeer AI - AI/ML Integration & Cloud Agents
• Riemann Computing Inc. - Distributed Computing Architecture
• Bleunomics - Bioinformatics & Drug Discovery Expertise
• Andrew Magdy Kamal - Project Lead & System Integration

Installation

Prerequisites

• C++ compiler (GCC 7+ or MSVC 2019+)
• CUDA Toolkit 11.0+ (for GPU acceleration)
• Python 3.8+
• Node.js 16+ (for the Decentralized Internet SDK)
• BOINC client/server software

Build Instructions

# Clone the repository
git clone https://huggingface.co/OpenPeerAI/DockingAtHOME
cd DockingAtHOME

# Initialize submodules
git submodule update --init --recursive

# Build the project
mkdir build && cd build
cmake ..
make -j$(nproc)

# Install
sudo make install

Getting Started

Installation

# Clone repository
git clone https://huggingface.co/OpenPeerAI/DockingAtHOME
cd DockingAtHOME

# Install dependencies
pip install -r requirements.txt
npm install

# Build C++/CUDA components
mkdir build && cd build
cmake .. && make -j$(nproc)

Quick Start with GUI

# Start the web-based GUI (fastest way to get started)
docking-at-home gui

# Or with Python
python -m docking_at_home.gui

# Open browser to http://localhost:8080

Quick Start Example (CLI)

from docking_at_home import DockingClient

# Initialize client (localhost mode)
client = DockingClient(mode="localhost")

# Submit docking job
job = client.submit_job(
    ligand="path/to/ligand.pdbqt",
    receptor="path/to/receptor.pdbqt",
    num_runs=100
)

# Monitor progress
status = client.get_status(job.id)

# Retrieve results
results = client.get_results(job.id)
print(f"Best binding energy: {results.best_energy} kcal/mol")

Running on Localhost

# Start server
docking-at-home server --port 8080

# In another terminal, run worker
docking-at-home worker --local

Python API

from docking_at_home.server import job_manager, initialize_server
import asyncio

async def main():
    await initialize_server()
    
    job_id = await job_manager.submit_job(
        ligand_file="molecule.pdbqt",
        receptor_file="protein.pdbqt",
        num_runs=100,
        use_gpu=True
    )
    
    # Monitor progress
    while True:
        job = job_manager.get_job(job_id)
        if job["status"] == "completed":
            print(f"Best energy: {job['results']['best_energy']}")
            break
        await asyncio.sleep(1)

asyncio.run(main())

Configuration

Configuration files are located in config/:

• autodock.conf - AutoDock parameters
• boinc_server.conf - BOINC server settings
• gpu_config.conf - CUDPP and GPU settings
• decentralized.conf - Distributed Network Settings
• cloud_agents.conf - AI orchestration parameters

Performance

On a typical configuration:

• CPU-only: ~100 docking runs/hour
• Single GPU (RTX 3090): ~2,000 docking runs/hour
• Distributed (1000 nodes): ~100,000+ docking runs/hour

Use Cases

• 🔬 Drug Discovery and Virtual Screening
• 🧪 Protein-Ligand Binding Studies
• 📚 Large-Scale Chemical Library Screening
• 🎓 Educational Computational Chemistry
• 🌍 Pandemic Response (e.g., COVID-19 drug discovery)

Contributing

We welcome contributions! Please see CONTRIBUTING.md for guidelines.

License

This project is licensed under the GNU General Public License v3.0 - see LICENSE for details.

Individual components retain their original licenses:

• AutoDock: GNU GPL v2
• BOINC: GNU LGPL v3
• CUDPP: BSD License

Citation

If you use Docking@HOME in your research, please cite:

@software{docking_at_home_2025,
  title={Docking@HOME: A Distributed Platform for Molecular Docking},
  author={OpenPeer AI and Riemann Computing Inc. and Bleunomics and Andrew Magdy Kamal},
  year={2025},
  url={https://huggingface.co/OpenPeerAI/DockingAtHOME}
}

Support

• 📧 Email: [email protected]
• � Issues: HuggingFace Issues
• 🤗 Community: HuggingFace Discussions

Acknowledgments

• The AutoDock development team at The Scripps Research Institute
• BOINC project at UC Berkeley
• CUDPP developers
• Lonero Team for the Decentralized Internet SDK
• OpenPeer AI for Cloud Agents framework

---

Made with ❤️ by the open-source computational chemistry community