plinder.core.structure.atoms

plinder.core.structure.atoms.biotite_ciffile() → TextFile

plinder.core.structure.atoms.atom_array_from_cif_file(structure: Path | AtomArray | AtomArrayStack, use_author_fields: bool = True) → AtomArray | None

plinder.core.structure.atoms.write_cif(arr: AtomArray, filepath: Path) → None

Write an AtomArray to an mmCIF file whose data block is the file stem.

plinder.core.structure.atoms.apply_mask(atoms: AtomArray | AtomArrayStack, mask: NDArray[bool]) → AtomArray | AtomArrayStack

Filter the atoms of an AtomArray or AtomArrayStack with a boolean mask.

plinder.core.structure.atoms.resn2seq(resn: Iterable[str]) → str

Convert residue names to one-letter code, with X for unknown names.

Selenocysteine and pyrrolysine become cysteine and lysine because biotite’s protein alphabet has no U or O.

plinder.core.structure.atoms.align_sequences(ref_seq: str, subject_seq: str, ref_numbering: Sequence[int] | None = None, subject_numbering: Sequence[int] | None = None) → tuple[str, str, list[int], list[int]]

Globally align two sequences and return their matched residues.

The alignment uses BLOSUM62 with affine gap penalties and no terminal penalty. Only positions where both sequences carry a known residue are kept, and returned as the two mapped sequences plus their residue numbering (1-based unless numbering is given).

plinder.core.structure.atoms.get_residue_index_mapping_mask(ref_seqs: dict[str, str], subject_arr: AtomArray | AtomArrayStack) → dict[str, NDArray[float64]]

Mark which reference-sequence positions are resolved in each chain.

plinder.core.structure.atoms.get_per_residue_mask(residue_reference_atom_list: Sequence[str], atom_list: Sequence[str]) → list[int]

plinder.core.structure.atoms.get_per_residue_atoms(atom_array: AtomArray | AtomArrayStack, resi: int, resn: str) → NDArray[str_]

plinder.core.structure.atoms.make_atom_mask(atom_array: AtomArray | AtomArrayStack, seq_res: str, seq_mask: Sequence[int | float]) → list[int]