o *j, @s dZddlZddlZddlmZmZmZddlZddl m Z ddl m Z ee ejfZee efZdZeeZejddGd d d Zdd e d ee d efddZd e fddZded e fddZded ejfddZ ddededeejd efddZ ddedeejd efddZdS)zProtein data type.N)AnyMappingOptional) PDBParser)residue_constantsZ>ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789T)frozenc@sVeZdZUdZejed<ejed<ejed<ejed<ejed<ejed<dd Zd S) Proteinz!Protein structure representation.atom_positionsaatype atom_mask residue_index chain_index b_factorscCs(tt|jtkrtdtddS)Nz(Cannot build an instance with more than z6 chains because these cannot be written to PDB format.)lennpuniquer PDB_MAX_CHAINS ValueError)selfro/var/www/html/Deteccion_Ine/venv/lib/python3.10/site-packages/modelscope/models/science/unifold/data/protein.py __post_init__<s  zProtein.__post_init__N)__name__ __module__ __qualname____doc__rndarray__annotations__rrrrrr s        rpdb_strchain_idreturnc st|}tdd}|d|}t|}t|dkr&tdt|d|d}g}g}g} g} g} g} |D]} |durD| j|krDq8| D]}|jd d kr^td | jd |jdd t j |j d}t j |t j}tt jdf}tt jf}tt jf}|D]#}|jt jvrq|j|t j|j<d|t j|j<|j|t j|j<qt|dkrqF||||| || |jd| | j| |qFq8t| }ddt|Dtfdd| D}tt|t| t|t| |t| dS)aTakes a PDB string and constructs a Protein object. WARNING: All non-standard residue types will be converted into UNK. All non-standard atoms will be ignored. Args: pdb_str: The contents of the pdb file chain_id: If chain_id is specified (e.g. A), then only that chain is parsed. Otherwise all chains are parsed. Returns: A new `Protein` parsed from the pdb contents. T)ZQUIETnonez,Only single model PDBs are supported. Found z models.rN z(PDB contains an insertion code at chain z and residue index z. These are not supported.X??cSsi|]\}}||qSrr).0ncidrrr z#from_pdb_string..csg|]}|qSrr)r)r+Zchain_id_mappingrr sz#from_pdb_string..)r r r r r r)ioStringIOrZ get_structurelistZ get_modelsrridrZ restype_3to1getZresnameZ restype_order restype_numrZzerosZ atom_type_numname atom_typesZcoordZ atom_orderZbfactorsumappendr enumeratearrayr)rrZpdb_fhparserZ structuremodelsmodelr r r r chain_idsrchainresZ res_shortnameZ restype_idxposmaskZ res_b_factorsZatomZunique_chain_idsr rr.rfrom_pdb_stringCs~             rDcCs*d}|d|dd|dd|d|dS) NZTER<6>5z >3r$>1>4r) atom_indexZ end_resnameZ chain_namer Z chain_endrrr _chain_ends rKprotcs|tjdgfdd}tj}g}|j}|j}|j}|jtj }|j tj }|j } t |tj kr7tdi} t|D]} | tkrLtdtdt| | | <q>|dd} |d } t|jd D]} | || kr|t| ||| d| || d|| d|| } | d7} ||| }t||| || | | D]n\}}}}|d krqd }t|d kr|nd |}d}d}d}|d }d}|d| dd |d|d|dd | || d|| d|dd|d d|dd|dd|d|dd|d|d}||| d7} qqe|t| ||d| |d|d|d|ddd |D}d!|d!S)"zConverts a `Protein` instance to a PDB string. Args: prot: The protein to convert to PDB. Returns: PDB string. r%cstj|dS)NZUNK)rZ restype_1to3r4)rrestypesrrres_1to3szto_pdb..res_1to3zInvalid aatypes.z The PDB format supports at most z chains.z MODEL 1r"rr(ATOMr$r'rErFz<4rHrGrIz z>8.3fr#z>6.2fz z>2ZENDMDLENDcSsg|]}|dqS)P)ljust)r)linerrrr/r-zto_pdb.. )rrOr7r r r r ZastyperZint32r ranyr5rrr PDB_CHAIN_IDSr9rangeshaperKziprjoin)rLrPr7Z pdb_linesr r r r r rr?irJZlast_chain_indexZ res_name_3Z atom_namerBrCZb_factorZ record_typer6Zalt_locZinsertion_codeZ occupancyelementZchargeZ atom_linerrNrto_pdbs              rbcCs tj|jS)asComputes an ideal atom mask. `Protein.atom_mask` typically is defined according to the atoms that are reported in the PDB. This function computes a mask according to heavy atoms that should be present in the given sequence of amino acids. Args: prot: `Protein` whose fields are `numpy.ndarray` objects. Returns: An ideal atom mask. )rZSTANDARD_ATOM_MASKr )rLrrrideal_atom_masks rcfeaturesresultrcCsbd|vr |dd}nt|d}|durt|d}t|d|d|d|dd||dS) aAssembles a protein from a prediction. Args: features: Dictionary holding model inputs. fold_output: Dictionary holding model outputs. b_factors: (Optional) B-factors to use for the protein. Returns: A protein instance. asym_idr"r NZfinal_atom_maskZfinal_atom_positionsr r r r r r rrZ zeros_liker)rdrerr rrrfrom_predictions ricCsbd|vr |dd}nt|d}|durt|d}t|d|d|d|dd||dS) zAssembles a standard pdb from input atom positions & mask. Args: features: Dictionary holding model inputs. b_factors: (Optional) B-factors to use for the protein. Returns: A protein instance. rfr"r NZ all_atom_maskZall_atom_positionsr rgrh)rdrr rrr from_feature's  rj)N)r dataclassesr0typingrrrnumpyrZBio.PDBrZ&modelscope.models.science.unifold.datarstrrZ FeatureDictZ ModelOutputr[rr dataclassrrDrKrbrcrirjrrrrs>     "N` !