zendock

zendock is a focused Python package for the molecular docking workflow sketched in notebooks/zendock_colab_demo.ipynb.

This package provides the stack needed to perform molecular docking for scientific use, prioritizing access and ease of use.

The base package keeps imports lightweight. Install extras for ligand/receptor preparation, docking, and visualization workflows.

Demo

Try the interactive Google Colab notebook:

Open the Zendock demo in Colab

Philosophy

• Lightweight import by default, optional extras for full workflows
• Notebook prototype, package implementation
• uv for development
• pip works for end users

Install

Users

Install from PyPI with regular pip:

pip install zendock

Install the complete workflow stack with:

pip install "zendock[all]"

Targeted extras are also available:

pip install "zendock[prep]"  # ligand/receptor preparation
pip install "zendock[dock]"  # Vina docking
pip install "zendock[viz]"   # 3D/interaction visualization

Development with uv

Create an environment and install the package in editable mode:

uv venv
source .venv/bin/activate
uv pip install -e ".[all]"

Install developer tools too:

uv pip install -e ".[dev]"

Development Workflow

Typical loop:

uv venv
source .venv/bin/activate
uv pip install -e ".[all,dev]"
pytest
python -m build

Project Layout

src/zendock/
  __init__.py
  docking.py
  exceptions.py
  fetching.py
  ligand.py
  models.py
  receptor.py
  result_store.py
  results.py
  tools.py
  utils.py
  workspace.py
tests/
Docking.ipynb
pyproject.toml

Current Public API

from zendock import DockingRunner, Ligand, OutputLayout, Receptor

DockingRunner is the notebook-friendly facade. New code can pass an OutputLayout to centralize output directories and use the typed result objects cached on runner.last_result after docking.

After run_docking(), cached results can be inspected as a Polars table:

results = runner.display_docking_results(ligand_name="ligand", n_poses=5)
# Prints the pose table plus mean lRMSD and mean uRMSD for those five poses.

rmsd_means = runner.get_docking_rmsd_means(ligand_name="ligand", n_poses=5)
# One-row Polars table: mean lRMSD, mean uRMSD

Interaction outputs are generated from saved docking results. Export methods return paths, while display helpers render the generated files in notebooks (or print a fallback in non-notebook environments):

runner.display_2d_interaction_map(ligand_name="ligand")
runner.display_3d_interaction_map(ligand_name="ligand")
runner.display_interaction_table(ligand_name="ligand")

The equivalent export methods are export_2d_interaction_diagram, export_3d_interaction_diagram, and export_interaction_table.

Notes

• Polars is used for tabular docking-result display.
• The package is licensed under MIT.