from phaser import residue_substitution from phaser import residue_topology as topology from phaser import rsam from iotbx.pdb import amino_acid_codes # Common macromolecule types PROTEIN_THREE_LETTER = ( amino_acid_codes.one_letter_given_three_letter_modified_aa.keys() + amino_acid_codes.one_letter_given_three_letter.keys() ) NUCLEIC_THREE_LETTER = ( " DA", # Adenine " DC", # Cytosine " DG", # Guanine " DT", # Thymine ) PROTEIN_ONE_LETTER = ( amino_acid_codes.one_letter_given_three_letter_modified_aa.values() + amino_acid_codes.one_letter_given_three_letter.values() ) NUCLEIC_ONE_LETTER = ( "A", # Adenine "C", # Cytosine "G", # Guanine "T", # Thymine ) PROTEIN_MAINCHAIN_ATOMS = ( " N ", " CA ", " C ", " O ", " OXT", ) PROTEIN_COMMON_RESIDUE_SIDECHAIN_TOPOLOGY_FOR = { "A": topology.ALA, # ALA "C": topology.CYS, # CYS "D": topology.ASP, # ASP "E": topology.GLU, # GLU "F": topology.PHE, # PHE "G": topology.GLY, # GLY "H": topology.HIS, # HIS "I": topology.ILE, # ILE "K": topology.LYS, # LYS "L": topology.LEU, # LEU "M": topology.MET, # MET "N": topology.ASN, # ASN "P": topology.PRO, # PRO "Q": topology.GLN, # GLN "R": topology.ARG, # ARG "S": topology.SER, # SER "T": topology.THR, # THR "V": topology.VAL, # VAL "W": topology.TRP, # TRP "Y": topology.TYR, # TYR } PROTEIN_UNKNOWN_TOPOLOGY = ( ( " CB ", () ), ) PROTEIN_SIDECHAIN_TOPOLOGIES = [ PROTEIN_COMMON_RESIDUE_SIDECHAIN_TOPOLOGY_FOR.get( resname, PROTEIN_UNKNOWN_TOPOLOGY ) for resname in PROTEIN_ONE_LETTER ] PROTEIN_MAINCHAIN_VAN_DER_WAALS_RADII = ( 1.35, # N 1.50, # CA 1.50, # C 1.34, # O 1.34, # OXT ) PROTEIN_COMMON_SIDECHAIN_VAN_DER_WAALS_RADII_FOR = { "A": { " CB ": 1.70, }, "C": { " CB ": 1.70, " SG ": 1.70, }, "D": { " CB ": 1.70, " CG ": 1.70, " OD1": 1.35, " OD2": 1.35, }, "E": { " CB ": 1.70, " CG ": 1.70, " CD ": 1.50, " OE1": 1.35, " OE2": 1.35, }, "F": { " CB ": 1.70, " CG ": 1.50, " CD1": 1.50, " CD2": 1.50, " CE1": 1.50, " CE2": 1.50, " CZ ": 1.50, }, "G": {}, "H": { " CB ": 1.70, " CG ": 1.70, " ND1": 1.35, " CD2": 1.50, " CE1": 1.50, " NE2": 1.35, }, "I": { " CB ": 1.70, " CG1": 1.70, " CG2": 1.70, " CD1": 1.70, }, "K": { " CB ": 1.70, " CG ": 1.70, " CD ": 1.70, " CE ": 1.70, " NZ ": 1.35, }, "L": { " CB ": 1.70, " CG ": 1.70, " CD1": 1.70, " CD2": 1.70, }, "M": { " CB ": 1.70, " CG ": 1.70, " SD ": 1.80, " CE ": 1.70, }, "N": { " CB ": 1.70, " CG ": 1.70, " OD1": 1.35, " ND2": 1.35, }, "P": { " CB ": 1.70, " CG ": 1.70, " CD ": 1.45, }, "Q": { " CB ": 1.70, " CG ": 1.70, " CD ": 1.50, " OE1": 1.35, " NE2": 1.35, }, "R": { " CB ": 1.70, " CG ": 1.70, " CD ": 1.70, " NE ": 1.35, " CZ ": 1.50, " NH1": 1.35, " NH2": 1.35, }, "S": { " CB ": 1.70, " OG ": 1.35, }, "T": { " CB ": 1.70, " OG1": 1.35, " CG2": 1.70, }, "V": { " CB ": 1.70, " CG1": 1.70, " CG2": 1.70, }, "W": { " CB ": 1.70, " CG ": 1.50, " CD1": 1.50, " CD2": 1.50, " NE1": 1.35, " CE2": 1.50, " CE3": 1.50, " CZ2": 1.50, " CZ3": 1.50, " CH2": 1.50, }, "Y": { " CB ": 1.70, " CG ": 1.50, " CD1": 1.50, " CD2": 1.50, " CE1": 1.50, " CE2": 1.50, " CZ ": 1.50, " OH ": 1.35, }, } PROTEIN_UNKNOWN_SIDECHAIN_VAN_DER_WAALS_RADII = {} PROTEIN_SIDECHAIN_VAN_DER_WAALS_RADII = [ PROTEIN_COMMON_SIDECHAIN_VAN_DER_WAALS_RADII_FOR.get( resname, PROTEIN_UNKNOWN_SIDECHAIN_VAN_DER_WAALS_RADII ) for resname in PROTEIN_ONE_LETTER ] PROTEIN_COMMON_RESIDUE_WEIGHT_FOR = { "A": 89.09, # Ala "C": 121.16, # Cys "D": 133.10, # Asp "E": 147.13, # Glu "F": 165.19, # Phe "G": 75.07, # Gly "H": 155.15, # His "I": 131.17, # Ile "K": 146.19, # Lys "L": 131.17, # Leu "M": 149.21, # Met "N": 132.12, # Asn "P": 115.13, # Pro "Q": 146.14, # Gln "R": 174.20, # Arg "S": 105.09, # Ser "T": 119.12, # Thr "V": 117.15, # Val "W": 204.23, # Trp "Y": 181.19, # Tyr } PROTEIN_UNKNOWN_RESIDUE_WEIGHT = 0 PROTEIN_RESIDUE_WEIGHTS = [ PROTEIN_COMMON_RESIDUE_WEIGHT_FOR.get( resname, PROTEIN_UNKNOWN_RESIDUE_WEIGHT ) for resname in PROTEIN_ONE_LETTER ] PROTEIN_COMMON_RESIDUE_H_COUNT_FOR = { "A": 7, # Ala "C": 7, # Cys "D": 7, # Asp "E": 9, # Glu "F": 11, # Phe "G": 5, # Gly "H": 9, # His "I": 13, # Ile "K": 14, # Lys "L": 13, # Leu "M": 11, # Met "N": 8, # Asn "P": 9, # Pro "Q": 10, # Gln "R": 14, # Arg "S": 7, # Ser "T": 9, # Thr "V": 11, # Val "W": 12, # Trp "Y": 11, # Tyr } PROTEIN_UNKNOWN_RESIDUE_H_COUNT = 0 PROTEIN_RESIDUE_H_COUNTS = [ PROTEIN_COMMON_RESIDUE_H_COUNT_FOR.get( resname, PROTEIN_UNKNOWN_RESIDUE_H_COUNT ) for resname in PROTEIN_ONE_LETTER ] class TopologyCache(object): """ Caches topology-related instances for quick access """ def __init__(self): self.sidechain_atom_for = {} self.mainchain_distance_for = {} def get_sidechain_atom_for(self, sidechain): return self.sidechain_atom_for.setdefault( sidechain, topology.SidechainAtom( topology = sidechain ) ) def get_mainchain_distance_for(self, sidechain): return self.mainchain_distance_for.setdefault( sidechain, topology.MainchainDistance( topology = sidechain, names = self.get_sidechain_atom_for( sidechain = sidechain ).names ) ) def get_mapping_between(self, source, target): return topology.SidechainCorrespondence( source = source, target = target ) # Class to encapsulate macromolecule-specific information class MacromoleculeType(object): """ Provides access to macromolecule-specific data """ def __init__( self, name, one_letter_codes, three_letter_codes, molecular_weights, hydrogen_counts, consecutivity, mainchain_atom_names, mainchain_atom_radii, sidechain_topologies, sidechain_atom_radii_dicts, secondary_structure_analyser, similarity_matrix_suite, min_length, sidechain_completion = [], unknown_one_letter = "X", unknown_three_letter = "UNK", unknown_molecular_weight = 0, unknown_hydrogen_count = 0, unknown_sidechain_topology = (), unknown_atom_radius = 1.70, condensation_weight_loss = 0, specific_volume = None, ): self.name = name self.one_letter_codes = one_letter_codes self.three_letter_codes = three_letter_codes self.molecular_weights = molecular_weights self.hydrogen_counts = hydrogen_counts self.condensation_weight_loss = condensation_weight_loss self.specific_volume = specific_volume self.consecutivity = consecutivity self.mainchain_atom_names = mainchain_atom_names self.mainchain_atom_radii = mainchain_atom_radii self.sidechain_topologies = sidechain_topologies self.sidechain_atom_radii_dicts = sidechain_atom_radii_dicts self.secondary_structure_analyser = secondary_structure_analyser self.similarity_matrix_suite = similarity_matrix_suite self.min_length = min_length self.sidechain_completion = sidechain_completion self.unknown_one_letter = unknown_one_letter self.unknown_three_letter = unknown_three_letter self.unknown_molecular_weight = unknown_molecular_weight self.unknown_hydrogen_count = unknown_hydrogen_count self.unknown_sidechain_topology = unknown_sidechain_topology self.unknown_atom_radius = unknown_atom_radius # Topology cache self.topology_cache = TopologyCache() def one_letter_for(self): return dict( zip( self.three_letter_codes, self.one_letter_codes ) ) def three_letter_for(self): return dict( zip( self.one_letter_codes, self.three_letter_codes ) ) def recognize_chain_type(self, chain, confidence = 0.9): residues = chain.residue_groups() if len( residues ) < self.min_length: return False all_count = len( residues ) known_codes = set( self.three_letter_codes ) known_count = len( [ rg for rg in residues if rg.atom_groups()[0].resname in known_codes ] ) return confidence * all_count <= known_count PROTEIN = MacromoleculeType( name = "protein", one_letter_codes = PROTEIN_ONE_LETTER, three_letter_codes = PROTEIN_THREE_LETTER, molecular_weights = PROTEIN_RESIDUE_WEIGHTS, hydrogen_counts = PROTEIN_RESIDUE_H_COUNTS, consecutivity = rsam.StructuralConsecutivity( pre_bond_atom = PROTEIN_MAINCHAIN_ATOMS[2], post_bond_atom = PROTEIN_MAINCHAIN_ATOMS[0], distance_cutoff = 1.70 ), mainchain_atom_names = PROTEIN_MAINCHAIN_ATOMS, mainchain_atom_radii = PROTEIN_MAINCHAIN_VAN_DER_WAALS_RADII, sidechain_topologies = PROTEIN_SIDECHAIN_TOPOLOGIES, sidechain_atom_radii_dicts = PROTEIN_SIDECHAIN_VAN_DER_WAALS_RADII, secondary_structure_analyser = rsam.DSSPInterface, similarity_matrix_suite = residue_substitution.protein, min_length = 2, sidechain_completion = [ rsam.ProteinSidechainCBCompletion ], unknown_molecular_weight = PROTEIN_UNKNOWN_RESIDUE_WEIGHT, unknown_hydrogen_count = PROTEIN_UNKNOWN_RESIDUE_H_COUNT, unknown_sidechain_topology = PROTEIN_UNKNOWN_TOPOLOGY, condensation_weight_loss = 18, specific_volume = 0.74, ) KNOWN = [ PROTEIN ] SOLVENT = MacromoleculeType( name = "solvent", one_letter_codes = ( "X", "X", "X" ), three_letter_codes = ( "HOH", "WAT", "0WO" ), molecular_weights = (), hydrogen_counts = (), consecutivity = lambda left, right: False, mainchain_atom_names = (), mainchain_atom_radii = (), sidechain_topologies = (), sidechain_atom_radii_dicts = (), secondary_structure_analyser = None, similarity_matrix_suite = residue_substitution.unknown, min_length = 0 ) UNKNOWN = MacromoleculeType( name = "unknown", one_letter_codes = (), three_letter_codes = (), molecular_weights = (), hydrogen_counts = (), consecutivity = lambda left, right: False, mainchain_atom_names = (), mainchain_atom_radii = (), sidechain_topologies = (), sidechain_atom_radii_dicts = (), secondary_structure_analyser = None, similarity_matrix_suite = residue_substitution.unknown, min_length = 0 ) def determine(chain, confidence = 0.9): for mmt in KNOWN: if mmt.recognize_chain_type( chain = chain, confidence = confidence ): return mmt return UNKNOWN