molsysmt.physchem.get_autodock_atom_types#

molsysmt.physchem.get_autodock_atom_types(molecular_system, typing_scheme, selection='all', structure_indices='all', chemical_state='reference', method='chemical_environment', return_report=False, syntax='MolSysMT', skip_digestion=False)[source]#

Getting full-graph chemical AutoDock labels before filtering source atoms.

Parameters:
  • molecular_system (molecular system) – Any supported form supplying a complete closed-shell covalent graph, elements, formal charges and supported bond orders. All H must be indexed. Missing aromaticity is perceived on a detached graph; known conflicts fail.

  • typing_scheme (str) – Required explicit ‘autodock4’ label vocabulary. Labels remain separate from MolSysMT atom_type, which denotes chemical element symbols.

  • selection (str, list, tuple or numpy.ndarray, default='all') – Atoms to return, after full-source classification. Indices are sorted and deduplicated; selected fragments are not classified in isolation.

  • structure_indices (int, list, tuple or numpy.ndarray, default='all') – Structure indices (0-based) for state resolution and spatial selection. Chemical typing itself does not use coordinates.

  • chemical_state (str or int, default='reference') – Selected state; ‘structure’ must resolve unambiguously across frames.

  • method (str, default='chemical_environment') – Experimental chemical_environment@1 profile, using element, explicit aromatic perception, valence, adjacency and formal charge. Later specific rules override defaults; no atom/residue name heuristic is used.

  • return_report (bool, default=False) – Return atom_ff_type and detached auditable report when True.

  • syntax (str, default='MolSysMT') – Syntax used for atom selection.

  • skip_digestion (bool, default=False) – Whether to skip MolSysMT’s internal argument digestion mechanism.

Returns:

U2 atom_ff_type labels with shape (n_selected_atoms,), including (0,). The optional report contains source indices, full evaluated scope, state, scheme/version, original software, ordered rules, per-atom winner codes, overlap counts and separate polar/nonpolar H indices. Indices and labels have no physical units. The source is unchanged.

Return type:

numpy.ndarray or dict

Raises:
  • ArgumentError – If the scheme, method, state, selection or frame request is invalid.

  • StructuralInconsistencyError – If chemistry is incomplete/contradictory, H are virtual, an element or charge environment has no qualified rule, or perception fails.

  • ImportError – If the required optional RDKit provider is unavailable.

Notes

The bounded profile supports H/C/N/O/F/P/S/Cl/Br/I with documented formal charge environments. It distinguishes aromatic C, trivalent conjugated N, positive N, two-connected nonaromatic S and H bonded to N/O/F/P/S. These AutoDock labels are a parameter profile, not universal donor/acceptor chemistry. See the User Guide for exact precedence and exclusions. Meeko 0.8.0 was a source reference; no Meeko code/table is vendored and it is not an installed dependency. Metal/pseudoatom/macrocycle typing, partial charges, protonation and H merging are outside this operation. RDKit performs explicit aromatic perception; optional Ackredit failures preserve results.

See also

molsysmt.build.assign_autodock_atom_types

Attach a full named type assignment.

molsysmt.physchem.get_aromaticity

Perceive the reusable aromatic model.

molsysmt.element.atom.get_atom_type_from_atom_ff_type

Decode existing labels.

Examples

>>> import molsysmt as msm
>>> result = msm.physchem.get_autodock_atom_types(
...     msm.systems['caffeine']['caffeine.sdf'], typing_scheme='autodock4')
>>> result.shape, result.dtype.name
((24,), 'str64')

User guide

See Getting AutoDock types for rules, provenance and limitations.

Added in version 1.0.0.