qcinf

Cheminformatics algorithms and structure utilities for qcdata structures.

The QC Suite

These packages work together to make quantum chemistry fast, structured, and interoperable.

• qcref - Reference data for computational chemistry: CODATA physical constants, conversion factors, a periodic table, and solvent properties, each value with its source.
• qcdata - Data structures for quantum chemistry inputs and results, with Jupyter visualizations. Documentation
• qcinf - Cheminformatics algorithms and structure utilities for qcdata structures.
• qccodec - Translates qcdata inputs into native program inputs and parses program outputs back into qcdata objects, one codec per program.
• qccompute - Runs quantum chemistry programs on qcdata inputs using qccodec codecs and returns structured results. Documentation
• BigChem - Distributed execution of qccompute calculations across clusters of computers or the cloud.
• ChemCloud - A web application and Python client that expose a BigChem cluster securely over the internet.

Installation

python -m pip install qcinf

To install with numba-accelerated kernels:

python -m pip install 'qcinf[fast]'

Additional backends can be added for conversion and compatibility features, including rdkit, openbabel, or install the whole suite with all:

python -m pip install 'qcinf[rdkit,openbabel]'

Or

python -m pip install 'qcinf[all]'

snapRMSD

snapRMSD is the native qcinf algorithm for symmetry-aware molecular RMSD. It uses bonded connectivity to find a connectivity-preserving atom assignment between two isomorphic structures, optionally aligns the structures, and returns the RMSD for the best assignment.

from qcdata import Structure
from qcinf import snap_rmsd

reference = Structure(
    symbols=["C", "H", "H", "H", "H"],
    geometry=[
        [0.0, 0.0, 0.0],
        [1.0, 1.0, 1.0],
        [-1.0, -1.0, 1.0],
        [-1.0, 1.0, -1.0],
        [1.0, -1.0, -1.0],
    ],
    connectivity=[(0, 1, 1.0), (0, 2, 1.0), (0, 3, 1.0), (0, 4, 1.0)],
)

candidate = Structure(
    symbols=["C", "H", "H", "H", "H"],
    geometry=[
        [0.0, 0.0, 0.0],
        [-1.0, 1.0, -1.0],
        [1.0, -1.0, -1.0],
        [1.0, 1.0, 1.0],
        [-1.0, -1.0, 1.0],
    ],
    connectivity=[(0, 1, 1.0), (0, 2, 1.0), (0, 3, 1.0), (0, 4, 1.0)],
)

rmsd = snap_rmsd(reference, candidate)

For the full result, use snap_rmsd_align_assign:

from qcinf import snap_rmsd_align_assign

rmsd, aligned_reference, assignment = snap_rmsd_align_assign(reference, candidate)

assignment[i] is the atom index in candidate matched to atom i in reference. The returned aligned_reference is a copy of reference with its geometry aligned to candidate.

Common options:

from qcinf import snap_align, snap_assign, snap_rmsd

# Compare structures without rotational/translational alignment.
rmsd = snap_rmsd(reference, candidate, align=False)

# Increase factoring depth for larger repeated peripheral groups.
rmsd = snap_rmsd(reference, candidate, factor_depth=2)

# Use the QCP alignment backend. Install with `qcinf[fast]` for numba acceleration.
rmsd = snap_rmsd(reference, candidate, alignment_backend="qcp")

# Return only the aligned structure or atom assignment.
aligned = snap_align(reference, candidate)
assignment = snap_assign(reference, candidate)

If a Structure does not include bonded connectivity, snapRMSD will infer it from covalent radii. For production workflows, pass validated connectivity on the structures or through a_connectivity= and b_connectivity= so the graph used for assignment is explicit and reproducible.

Contributing

See CONTRIBUTING.md for development setup, the pre-commit workflow, running tests, and the release process.

Support

If you have any issues with qcinf or would like to request a feature, please open an issue.