DockingAtHOME / CHANGELOG.md
Mentors4EDU's picture
Upload 42 files
35aaa09 verified
|
raw
history blame
2.87 kB

Changelog

All notable changes to the Docking@HOME project will be documented in this file.

The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.

[1.0.0] - 2025-11-19

Added

Core Features

  • AutoDock 4.2.6 integration for molecular docking
  • CUDA/CUDPP GPU acceleration for parallel docking
  • BOINC distributed computing framework integration
  • The Decentralized Internet SDK for Distributed Network Settings-based coordination
  • Cloud Agents AI-powered task orchestration
  • HuggingFace model card and integration

Components

  • C++ BOINC wrapper with client/server support
  • CUDA kernels for GPU-accelerated docking
  • Genetic algorithm implementation on GPU
  • JavaScript decentralized coordinator
  • Python Cloud Agents orchestrator
  • Command-line interface (CLI)
  • Python API

Build System

  • CMake build configuration
  • Python package setup
  • Node.js package configuration
  • Cross-platform support

Documentation

  • Comprehensive README with architecture diagrams
  • HuggingFace Model Card
  • Contributing guidelines
  • License (GPL-3.0)
  • Example workflows
  • Configuration guides

Features

  • Task submission and tracking
  • Real-time progress monitoring
  • Result retrieval and analysis
  • GPU benchmarking
  • Worker node management
  • System statistics
  • Auto-scaling recommendations

Authors

  • 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

Technical Specifications

  • Support for PDBQT format (ligands and receptors)
  • GPU acceleration with CUDA
  • Distributed computing via BOINC
  • Distributed Network Settings coordination via the Decentralized Internet SDK
  • AI optimization via Cloud Agents
  • Performance: ~2,000 runs/hour on single RTX 3090
  • Distributed: 100,000+ runs/hour on 1000 nodes

Known Limitations

  • Requires CUDA-capable GPU for optimal performance
  • Limited receptor flexibility (rigid docking)
  • Simplified solvation models
  • Requires external validation of results

Future Releases

[1.1.0] - Planned

  • Enhanced flexibility modeling
  • Improved solvation models
  • Web-based user interface
  • Real-time visualization
  • Enhanced metal coordination handling

[2.0.0] - Planned

  • Machine learning scoring functions
  • Multi-receptor ensemble docking
  • Enhanced Cloud Agents integration
  • Advanced distributed network features
  • Native cloud deployment support

Note: For detailed changes in each release, see the HuggingFace Releases page.