denspart

Official Resources

• Homepage: https://github.com/theochem/denspart
• GitHub: https://github.com/theochem/denspart
• Documentation: GitHub README
• License: GNU General Public License v3.0

Overview

denspart is a Python package for atoms-in-molecules density partitioning. It implements various stockholder-type partitioning schemes including Iterative Stockholder Atoms (ISA), Minimal Basis Iterative Stockholder (MBIS), and Gaussian Iterative Stockholder (GISA) methods for computing atomic charges and multipole moments.

Scientific domain: Charge partitioning, atoms-in-molecules, stockholder methods Target user community: Computational chemists developing force fields and analyzing charge distributions

Theoretical Methods

• Iterative Stockholder Atoms (ISA)
• Minimal Basis Iterative Stockholder (MBIS)
• Gaussian Iterative Stockholder (GISA)
• Multipole moment calculation
• Stockholder partitioning

Capabilities (CRITICAL)

• ISA Method: Iterative stockholder partitioning
• MBIS Method: Minimal basis stockholder
• GISA Method: Gaussian basis stockholder
• Multipole Moments: Full multipole expansion
• Python Native: Modern Python implementation
• NumPy Based: Efficient array operations

Sources: denspart GitHub repository, theochem group

Key Strengths

Modern Implementation:

• Python 3 native
• NumPy/SciPy based
• Clean API design
• Active development

Multiple Methods:

• ISA, MBIS, GISA
• Consistent interface
• Method comparison
• Flexibility

theochem Ecosystem:

• IOData integration
• Grid library support
• ChemTools compatible
• Consistent tools

Inputs & Outputs

• Input formats:

  • NumPy .npz density files
  • Grid-based density data
  • Via IOData parsers

• Output data types:

  • Atomic charges
  • Multipole moments
  • Partitioned densities
  • NumPy arrays

Installation

pip install denspart
# Or from source
git clone https://github.com/theochem/denspart.git
cd denspart
pip install -e .

Usage Examples

from denspart.mbis import MBISProModel
import numpy as np

# Load density data
data = np.load('density.npz')

# Perform MBIS partitioning
pro_model = MBISProModel.from_geometry(atnums, atcoords)
pro_model.run(density, grid)

# Get atomic charges
charges = pro_model.charges
multipoles = pro_model.multipole_moments

Performance Characteristics

• Speed: Efficient NumPy operations
• Memory: Scales with grid size
• Convergence: Iterative methods

Limitations & Known Constraints

• Grid input: Requires pre-computed density grid
• Documentation: Evolving
• Molecular focus: Primarily for molecules
• Dependencies: Requires theochem ecosystem

Comparison with Other Tools

• vs DDEC: denspart open-source, different methods
• vs Hirshfeld: denspart implements iterative variants
• vs horton-part: Related but separate packages
• Unique strength: Modern Python implementation of ISA/MBIS

Application Areas

• Force field parameterization
• Charge distribution analysis
• Electrostatic property prediction
• Reactivity indices
• Machine learning features

Best Practices

• Use appropriate grid density
• Compare multiple methods
• Verify charge conservation
• Validate against known systems

Community and Support

• GitHub repository
• theochem research group
• GPL v3 licensed
• Active development

Verification & Sources

Primary sources:

1. GitHub: https://github.com/theochem/denspart
2. theochem group (McMaster University)

Confidence: VERIFIED

Verification status: ✅ VERIFIED

• GitHub repository: ACCESSIBLE
• Source code: OPEN (GPL-3.0)
• Developer: theochem group
• Method: ISA/MBIS implementation