Parser

The parser is the core entry point for converting structural and sequence files (mmCIF, PDB, FASTA, SMILES, etc.) into Biotite’s AtomArray API. It supports extensive options for annotation, filtering, and caching.

Example Usage

from atomworks.io.parser import parse
result = parse(filename="/databases/rcsb/cif/ne/3nez.cif.gz")
print(result["chain_info"])

Returned Dictionary

• chain_info: Mapping of chain IDs to sequence, type, and metadata

• ligand_info: Information about ligands in the structure

• asym_unit: AtomArrayStack of the asymmetric unit

• assemblies: Mapping of assembly IDs to AtomArrayStacks

• metadata: Structure-level metadata

• extra_info: Internal-use information for caching and compatibility

Parsing Arguments

Name	Type	Default	Description
filename	PathLike / io.StringIO / io.BytesIO	—	Path to the structural file. Supports .cif, .cif.gz, .pdb, etc.
add_missing_atoms	bool	True	Add missing atoms to the structure. Useful for unresolved residues. Also adds intra- and inter-residue bonds.
add_id_and_entity_annotations	bool	True	Add id and entity annotations at chain, pn-unit, and molecule level to the AtomArray.
add_bond_types_from_struct_conn	list[str]	[“covale”]	List of bond types to add from struct_conn. Default is only covalent bonds.
remove_ccds	list[str] or None	crystallization aids	CCD codes to remove from the structure.
remove_waters	bool	True	Remove water molecules from the structure.
fix_ligands_at_symmetry_centers	bool	True	Patch non-polymer residues at symmetry centers that clash with themselves.
fix_arginines	bool	True	Fix arginine naming ambiguity.
fix_formal_charges	bool	True	Fix formal charges on atoms involved in inter-residue bonds.
convert_mse_to_met	bool	False	Convert selenomethionine (MSE) to methionine (MET).
remove_hydrogens	bool or None	None	Remove hydrogens from structure. Deprecated; use hydrogen_policy instead.
hydrogen_policy	“keep” / “remove” / “infer”	“keep”	Whether to keep, remove, or infer hydrogens.
model	int or None	None	Model number for NMR entries.
build_assembly	“first” / “all” / list[str] / tuple[str] / None	“all”	Which assembly to build: None, “first”, “all”, or specific IDs.
extra_fields	list[str] / “all” / None	None	Extra fields to include in the AtomArrayStack.

Caching Arguments

Name	Type	Default	Description
load_from_cache	bool	False	Load pre-compiled results from cache.
save_to_cache	bool	False	Save parsed results to cache for faster future retrieval.
cache_dir	PathLike / None	None	Directory for cached results. Required if caching is enabled.