MolCrysKit: Molecular Crystal Toolkit

Overview

MolCrysKit is a Python toolkit designed for handling molecular crystals, providing utilities for parsing crystallographic data, identifying molecules within crystals, and performing various analyses on molecular crystals using graph theory and the Atomic Simulation Environment (ASE).

Key Features

• Robust Molecule Identification: Identify individual molecules within a crystal structure using graph-based algorithms
• Disorder Handling: Process disordered structures with graph algorithms
• Topological Surface Generation: Create surface slabs while preserving molecular topology
• Hydrogen Completion: Add hydrogen atoms with heuristic geometric placement rules
• Cluster Carving: Disorder resolution → desolvation → coordination-shell cluster extraction
• Nanocluster Carving: Build arbitrary finite shapes while preserving complete molecular or unit-cell topology
• LLM-based AI agent Friendliness: Try on https://matmaster.bohrium.com/matmaster or build your own skill for your agent

Molecule Identification

Disorder Handling

Topological Surface Generation

Hydrogen Completion

Cluster Carving

Shape Operations

Installation

From PyPI (recommended)

pip install molcrys-kit

From source (development)

git clone https://github.com/SchrodingersCattt/MolCrysKit.git
cd MolCrysKit
pip install -e ".[dev]"

All dependencies are declared in pyproject.toml (there is no separate requirements.txt). requires-python = ">=3.10". The available extras are:

So a contributor environment is pip install -e ".[dev]" and a CI / minimal test environment is pip install -e ".[test]".

Optional separate formatting commit

In an activated [dev] environment, run pre-commit install --hook-type pre-commit and python scripts/install_style_hook.py from the repository root. The first command installs the existing lint and file checks. The second installs an optional Git post-commit hook (and refuses to overwrite an existing one). After a commit, this hook runs Ruff format on Python files in that commit. If formatting changes them, it creates a separate style: format Python from previous commit commit. It skips formatting when tracked files have uncommitted changes, so it never incorporates unrelated work. Git worktrees of the same clone share the hooks directory; install this hook only if you want it available to that clone's worktrees. The usual Ruff format hook remains available manually with pre-commit run ruff-format --hook-stage manual --all-files.

Ruff formats whitespace and layout, but it does not rewrite dense logic into more readable steps. The follow-up commit only happens when formatting actually changes a file; a commit that is already formatted gets no extra commit.

Quick Start

Here's a simple example of how to use MolCrysKit:

import molcrys_kit as mck
from ase import Atoms

# 1. Create a toy system (e.g., 2 Water molecules in a unit cell)
# In practice, you would typically load this from a file: atoms = read('cif_file.cif')
atoms = Atoms(
    symbols=['O', 'H', 'H', 'O', 'H', 'H'],
    positions=[
        [1.0, 1.0, 1.0], [1.8, 1.0, 1.0], [0.7, 1.6, 1.0],  # Molecule 1
        [5.0, 5.0, 5.0], [5.8, 5.0, 5.0], [4.7, 5.6, 5.0]   # Molecule 2
    ],
    cell=[10.0, 10.0, 10.0],
    pbc=True
)

# 2. Initialize MolecularCrystal (Automatically identifies molecules via graph logic)
crystal = mck.MolecularCrystal.from_ase(atoms)

# 3. Access Crystal & Molecular Properties
print(f"Lattice Parameters: {crystal.get_lattice_parameters()}")
print(f"Identified Molecules: {len(crystal.molecules)}") 

mol = crystal.molecules[0]
print(f"Molecule 1 Formula: {mol.get_chemical_formula()}")
print(f"Molecule 1 Center of Mass: {mol.get_center_of_mass()}")

Citation

If you use MolCrysKit in academic work, please cite:

Guo, M.-Y.; Zhang, W.-X. MolCrysKit: A Topology-Aware Toolkit for Bridging Experimental Molecular-Crystal Structures and Simulation-Ready Modeling. J. Chem. Inf. Model. 2026, 66 (9), 4999-5007. https://doi.org/10.1021/acs.jcim.6c00168

For exact reproduction of the published JCIM results, use the archived v0.1.0 release together with the versioned container image and the material under paper/. The main branch may continue to evolve after publication.

Docker / Cloud

See Docker Guide for local Docker, Bohrium cloud deployment, and GHCR image publication.

Command Line Interface

Installing MolCrysKit also installs the mck command. The CLI is self-documenting; use --help at any level to see the exact arguments for your installed version:

mck --help
mck io --help
mck operate --help
mck analyze --help
mck operate cluster --help
mck operate nanocluster --help
mck operate void --help
mck build chain --help
mck validate-periodic-bundle --help

The command groups roughly mirror the Python package layout:

• mck io ... — inspect molecular-crystal contents, extract molecules, and convert structures (info, molecules, extract-molecule, convert).
• mck operate ... — generate modified structures (disorder, disorder-supercell, add-h, slab, cluster, nanocluster, void, supercell, vacancy, desolvate, interpolate, reorient).
• mck analyze ... — print analysis reports (summary, bfdh, interactions, bond-orders, polyhedra, sanity-check).
• mck build chain ... — construct a geometry-native periodic chain bundle.
• mck validate-periodic-bundle ... — validate a periodic bundle and its sidecar.

Documentation

molcrys_kit/ — source code · scripts/ — diagnostic utilities · examples/ — CIF structure files

License

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