Skip to content
bioscinetworkPublic

About

Professional, reproducible automation platform and Qt GUI for AutoDock Vina, AutoDock4, and AutoGrid4. Features rigid/flexible docking, interaction analysis, pose validation, and portable packaging.

Topics

Resources

Stars

0 stars

Watchers

0 watching

Forks

Latest commit

 

History

12 Commits

Folders and files

NameName
Last commit message
Last commit date
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

AutoDock Suite Platform (ADSP) v0.3.0

AutoDock Suite Platform (ADSP) is a research-oriented molecular docking workflow platform providing unified automation, preparation, docking execution, pose parsing, interaction analysis, complex building, and auditable reporting for AutoDock4 and AutoDock Vina.

ADSP is designed for reproducible computational chemistry and structural biology research, featuring hardware-accelerated PySide6 desktop GUI and comprehensive CLI automation.


Key Capabilities & Architecture

  • Dual-Engine Architecture: Independent, isolated execution pathways for AutoDock Vina (empirical scoring, Monte Carlo global optimization) and AutoDock4 (semi-empirical force field, Lamarckian Genetic Algorithm).
  • Strict Thermodynamic Isolation:
    • AutoDock Vina: Binding affinity strictly reported in kcal/mol. No synthetic thermodynamic inference ($K_i$, $\Delta H$, $\Delta S$ are never fabricated).
    • AutoDock4: Binding free energy and inhibition constant ($K_i$) are strictly DLG-driven from AutoDock4's internal thermodynamic analysis.
  • PDBQT & Structure Validation: Fixed-column format validation enforcing atom names, coordinates, partial charges, and atom type fields with Open Babel round-trip compatibility.
  • Graph-Constrained RMSD: Hungarian algorithm automorphism symmetry matching and Maximum Common Substructure (MCS) via RDKit/obrms. Zero coordinate-order fallback; structural mismatch raises explicit failures rather than reporting misleading 0.0 Å.
  • Auditable Reporting: Automated multi-format reports (CSV, XLSX, JSON, TXT, HTML) with complete provenance tracking (engine, hashes, parameters, timestamps).

Repository Structure

D:\Projects\ADSP
├── main.py                     # Unified CLI entry point
├── config.py                   # Project configuration and directory resolution
├── models.py                   # Canonical data schemas and provenance models
├── prepare.py                  # Macromolecule and ligand preparation engine
├── validators.py               # Fixed-column PDBQT and geometry validation
├── dlg_extract.py              # Canonical AutoDock4 DLG extraction engine
├── vina_workflow.py            # Vina docking workflow orchestration
├── autodock4_workflow.py       # AutoDock4 & AutoGrid4 workflow orchestration
├── reporting.py                # Multi-format report generation (CSV/XLSX/JSON/TXT)
├── interactions.py             # Protein-ligand interaction profiler (HB, vdW, pi)
├── complex_builder.py          # Receptor-ligand complex generation & validation
├── executables.py              # Portable runtime binary & toolchain resolution
│
├── gui_qt/                     # Canonical PySide6 (Qt6) hardware-accelerated GUI
├── gui/                        # Legacy/fallback desktop GUI and icon assets
│
├── tests/                      # Automated test suite (124 pytest tests)
│   ├── test_suite.py
│   ├── test_packaging.py
│   ├── test_ad4_compatibility.py
│   └── ...
│
├── benchmarks/                 # Curated crystallographic benchmark fixtures
│   ├── 1CA2/                   # Human Carbonic Anhydrase II (Zn²⁺)
│   ├── 1MBN/                   # Sperm Whale Myoglobin (Fe²⁺/HEM)
│   └── 2NV6/                   # M. tuberculosis InhA / INH-NAD adduct
│
├── data/                       # Structural data directories
│   ├── receptors/              # Target macromolecule files
│   ├── ligands/                # Compound libraries
│   └── reference/              # Reference ligand conformers (ADP, HRM, Donepezil)
│
├── parameter_profiles/         # Cryptographically verified parameter profiles
│   ├── ad4_standard_4.2/       # AD4_parameters.dat (SHA-256 verified)
│   └── ad4_1_bound/            # AD4.1_bound.dat (SHA-256 verified)
│
├── external/
│   └── autodock4zn/            # Isolated AutoDock4Zn specialized pipeline archive
│
├── bin/                        # Bundled Windows executables (Vina, AD4, AutoGrid4, Open Babel)
├── docs/                       # Comprehensive documentation (USER_GUIDE, DEVELOPER_GUIDE)
├── scripts/                    # Build, setup, and execution batch scripts
├── AutoDockSuitePro.spec       # Portable PyInstaller packaging specification
├── build_exe.py                # Standalone distribution packaging script
├── pyproject.toml              # Build system & packaging configuration
├── project_config.toml         # Portable configuration file
├── CITATION.cff                # Software citation metadata
├── LICENSE                     # GNU General Public License v2.0+
└── README.md

Quick Start

Prerequisites & Installation

ADSP requires Python 3.10+ (64-bit) on Windows.

  1. Clone or download the repository:

    git clone https://github.com/your-org/ADSP.git
    cd ADSP
  2. Install dependencies:

    scripts\install_dependencies.bat

    Or via pip:

    pip install -r requirements.txt

Launching the Graphical Interface

Double-click run_gui.bat or run:

python main.py --gui

CLI Docking Workflows

Initialize a project configuration:

python main.py --init

Validate environment, toolchains, and inputs:

python main.py --validate --config project_config.toml

Execute AutoDock Vina docking:

python main.py --config project_config.toml --engine VINA

Execute AutoDock4 docking:

python main.py --config project_config.toml --engine AUTODOCK4

Scientific Parameter Profiles

ADSP bundles authoritative AutoDock4 parameter profiles verified by SHA-256 digests:

Profile ID Version Parameter File Reference / Source SHA-256 Checksum
ad4_standard_4.2 4.2 AD4_parameters.dat AutoDock4 Force Field 625DE5779B914382E21A135C776EFBC02B4221085BD0280118D103CCDD93EA7C
ad4_1_bound 4.1-bound AD4.1_bound.dat Huey et al. (2007) J Comput Chem 6B98F7AB508F4882801938F8CED1C0BF38096496155A8005BAF941A201781CE8

AutoDock4Zn Status & Isolation

The AutoDock4Zn zinc coordination pipeline (external/autodock4zn/AutoDock4Zn-Pipeline-main.zip) is isolated from standard AD4 runs. Standard AD4 parameter files include non-bonded Zn and Fe atom types; they do not apply specialized coordination pseudo-atoms unless explicitly configured through specialized legacy pipelines.


Standalone Portable Packaging

ADSP can be compiled into a fully portable, zero-dependency Windows distribution:

python build_exe.py

This generates:

  • Standalone folder: dist\AutoDockSuitePro\
  • Portable ZIP: dist\AutoDockSuitePro_Portable.zip

The portable distribution contains all Python runtimes, Qt libraries, Open Babel binaries, Vina, AutoDock4, AutoGrid4, parameter profiles, and default configs. It runs out-of-the-box on clean Windows systems without Python installed and does not depend on developer paths.


Testing & Quality Assurance

Run the comprehensive automated test suite (124 tests):

pytest

Scientific Scope & Limitations

  1. Thermodynamics: Vina calculates empirical binding affinities (kcal/mol), not thermodynamic free energies ($\Delta G$). Inhibition constants ($K_i$) are only derived when executing AutoDock4 with real DLG outputs.
  2. Charges: Kollman partial charges assigned via ADSP fallback mode are structural approximations and should not be conflated with canonical quantum mechanical ESP charges.
  3. 2NV6 / INH-NAD: Crystallographic adduct extraction (ZID.cif / 2NV6 (2).pdb, 52 heavy atoms / 82 explicit atoms) is deliberately kept distinct from independent SDF models (2nv6_B_ZID.sdf). False redocking validation is never claimed.

Citation

If you use ADSP in academic research, please cite:

@software{adsp2026,
  author = {Chibuike Praise Okechukwu},
  title = {AutoDock Suite Platform (ADSP): Unified Molecular Docking Platform},
  version = {0.3.0},
  year = {2026},
  url = {https://github.com/bioscinetwork/ADSP}
}

Please also cite the underlying engine and toolchain publications:

  • AutoDock4: Morris et al. (2009), J Comput Chem 30:2785-2791.
  • AutoDock Vina: Trott & Olson (2010), J Comput Chem 31:455-461; Eberhardt et al. (2021), J Chem Inf Model 61:3891-3898.
  • Open Babel: O'Boyle et al. (2011), J Cheminform 3:33.
  • RDKit: RDKit: Open-source cheminformatics (https://www.rdkit.org).
  • Meeko: Forli Lab, Scripps Research (https://github.com/forlilab/Meeko).

License

ADSP is licensed under the GNU General Public License v2.0 or later (LICENSE). Bundled third-party binaries and dependencies retain their respective licenses (see LICENSE and .licenses/).

About

Professional, reproducible automation platform and Qt GUI for AutoDock Vina, AutoDock4, and AutoGrid4. Features rigid/flexible docking, interaction analysis, pose validation, and portable packaging.

Topics

Resources

Stars

0 stars

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages