def _write_domain_sequence_file(self, protein_ids, protein_domain_defs,
protein_sequences):
"""Write a fasta file containing target domain sequences."""
sequence_seqrecords = []
for protein_id, protein_domain_def, protein_sequence in zip(
protein_ids, protein_domain_defs, protein_sequences):
#
sequence_id = "{}.{}".format(protein_id, protein_domain_def)
domain_def = structure_tools.decode_domain_def(protein_domain_def,
merge=True,
return_string=False)
seqrecord = SeqRecord(
id=sequence_id,
name=protein_id,
seq=Seq(protein_sequence[domain_def[0] - 1:domain_def[1]]),
)
sequence_seqrecords.append(seqrecord)
sequence_filename = "_".join(seqrec.id
for seqrec in sequence_seqrecords)
sequence_file = op.join(conf.CONFIGS["unique_temp_dir"],
sequence_filename + ".fasta")
with open(sequence_file, "w") as ofh:
SeqIO.write(sequence_seqrecords, ofh, "fasta")
return sequence_file, sequence_seqrecords
def _write_domain_sequence_file(self, protein_ids, protein_domain_defs, protein_sequences):
"""Write a fasta file containing target domain sequences."""
sequence_seqrecords = []
for protein_id, protein_domain_def, protein_sequence in zip(
protein_ids, protein_domain_defs, protein_sequences
):
#
sequence_id = "{}.{}".format(protein_id, protein_domain_def)
domain_def = structure_tools.decode_domain_def(protein_domain_def, merge=True, return_string=False)
seqrecord = SeqRecord(
id=sequence_id, name=protein_id, seq=Seq(protein_sequence[domain_def[0] - 1 : domain_def[1]])
)
sequence_seqrecords.append(seqrecord)
sequence_filename = "_".join(seqrec.id for seqrec in sequence_seqrecords)
sequence_file = op.join(conf.CONFIGS["unique_temp_dir"], sequence_filename + ".fasta")
with open(sequence_file, "w") as ofh:
SeqIO.write(sequence_seqrecords, ofh, "fasta")
return sequence_file, sequence_seqrecords
def get_mutation_data(self):
""""""
d = self.d
uniprot_id_1 = self.uniprot_id
mutation = self.mutation
if isinstance(d, elaspic_database.UniprotDomain):
logger.debug("Analyzing core mutation for uniprot: %s" %
uniprot_id_1)
logger.debug("model_domain_def: {}".format(
d.template.model.model_domain_def))
domain_start, domain_end = structure_tools.decode_domain_def(
d.template.model.model_domain_def)
mutation_idx = 0
core_or_interface = "core"
domain_start = domain_start or 1
domain_end = domain_end or len(self.sequence.sequence)
if int(mutation[1:-1]) < domain_start or int(
mutation[1:-1]) > domain_end:
raise errors.MutationOutsideDomainError(
"Mutation {} falls outside the domain with definitions {}."
.format(mutation, d.template.model.model_domain_def))
elif isinstance(d, elaspic_database.UniprotDomainPair):
mutation_pos = int(mutation[1:-1])
interacting_aa_1 = self._get_interacting_aa(d, 1)
interacting_aa_2 = self._get_interacting_aa(d, 2)
assert uniprot_id_1 in [
d.uniprot_domain_1.uniprot_id,
d.uniprot_domain_2.uniprot_id,
]
logger.debug("sequence: {}".format(self.sequence.sequence))
if uniprot_id_1 == d.uniprot_domain_1.uniprot_id and mutation_pos in interacting_aa_1:
logger.debug("Mutation is inside the first domain.")
logger.debug("model_domain_def: {}".format(
d.template.model.model_domain_def_1))
domain_start, domain_end = structure_tools.decode_domain_def(
d.template.model.model_domain_def_1)
if domain_start is None:
domain_start = 1
mutation_idx = 0
core_or_interface = "interface"
domain_start = domain_start or 1 # if None
domain_end = domain_end or len(self.sequence.sequence)
elif uniprot_id_1 == d.uniprot_domain_2.uniprot_id and mutation_pos in interacting_aa_2:
logger.debug("Mutation is inside the second domain.")
logger.debug("model_domain_def: {}".format(
d.template.model.model_domain_def_2))
domain_start, domain_end = structure_tools.decode_domain_def(
d.template.model.model_domain_def_2)
mutation_idx = 1
core_or_interface = "interface"
domain_start = domain_start or 1
domain_end = domain_end or len(self.sequence.sequence)
else:
# Mutation is outside the interface
error_message = "Mutated residue {} not involved in the interaction!".format(
mutation[:-1])
logger.error(error_message)
logger.error("Uniprot ID: {}\tMutation: {}".format(
uniprot_id_1, mutation))
logger.error("Uniprot ID 1: {}\tInteracting AA 1: {}".format(
d.uniprot_domain_1.uniprot_id, interacting_aa_1))
logger.error("Uniprot ID 2: {}\tInteracting AA 2: {}".format(
d.uniprot_domain_2.uniprot_id, interacting_aa_2))
raise errors.MutationOutsideInterfaceError(error_message)
position_domain = int(mutation[1:-1]) - domain_start + 1
mutation_domain = mutation[0] + str(position_domain) + mutation[-1]
class MutationData:
def __init__(self, **kwargs):
self.__dict__ = kwargs
mut_data = MutationData(
mutation_idx=mutation_idx,
mutation_domain=mutation_domain,
core_or_interface=core_or_interface,
)
return mut_data
def get_mutation_data(self):
"""
"""
d = self.d
uniprot_id_1 = self.uniprot_id
mutation = self.mutation
if isinstance(d, elaspic_database.UniprotDomain):
logger.debug("Analyzing core mutation for uniprot: %s" % uniprot_id_1)
logger.debug("model_domain_def: {}".format(d.template.model.model_domain_def))
domain_start, domain_end = structure_tools.decode_domain_def(d.template.model.model_domain_def)
mutation_idx = 0
core_or_interface = "core"
domain_start = domain_start or 1
domain_end = domain_end or len(self.sequence.sequence)
if int(mutation[1:-1]) < domain_start or int(mutation[1:-1]) > domain_end:
raise errors.MutationOutsideDomainError(
"Mutation {} falls outside the domain with definitions {}.".format(
mutation, d.template.model.model_domain_def
)
)
elif isinstance(d, elaspic_database.UniprotDomainPair):
mutation_pos = int(mutation[1:-1])
interacting_aa_1 = self._get_interacting_aa(d, 1)
interacting_aa_2 = self._get_interacting_aa(d, 2)
assert uniprot_id_1 in [d.uniprot_domain_1.uniprot_id, d.uniprot_domain_2.uniprot_id]
logger.debug("sequence: {}".format(self.sequence.sequence))
if uniprot_id_1 == d.uniprot_domain_1.uniprot_id and mutation_pos in interacting_aa_1:
logger.debug("Mutation is inside the first domain.")
logger.debug("model_domain_def: {}".format(d.template.model.model_domain_def_1))
domain_start, domain_end = structure_tools.decode_domain_def(d.template.model.model_domain_def_1)
if domain_start is None:
domain_start = 1
mutation_idx = 0
core_or_interface = "interface"
domain_start = domain_start or 1 # if None
domain_end = domain_end or len(self.sequence.sequence)
elif uniprot_id_1 == d.uniprot_domain_2.uniprot_id and mutation_pos in interacting_aa_2:
logger.debug("Mutation is inside the second domain.")
logger.debug("model_domain_def: {}".format(d.template.model.model_domain_def_2))
domain_start, domain_end = structure_tools.decode_domain_def(d.template.model.model_domain_def_2)
mutation_idx = 1
core_or_interface = "interface"
domain_start = domain_start or 1
domain_end = domain_end or len(self.sequence.sequence)
else:
# Mutation is outside the interface
error_message = "Mutated residue {} not involved in the interaction!".format(mutation[:-1])
logger.error(error_message)
logger.error("Uniprot ID: {}\tMutation: {}".format(uniprot_id_1, mutation))
logger.error(
"Uniprot ID 1: {}\tInteracting AA 1: {}".format(d.uniprot_domain_1.uniprot_id, interacting_aa_1)
)
logger.error(
"Uniprot ID 2: {}\tInteracting AA 2: {}".format(d.uniprot_domain_2.uniprot_id, interacting_aa_2)
)
raise errors.MutationOutsideInterfaceError(error_message)
position_domain = int(mutation[1:-1]) - domain_start + 1
mutation_domain = mutation[0] + str(position_domain) + mutation[-1]
class MutationData:
def __init__(self, **kwargs):
self.__dict__ = kwargs
mut_data = MutationData(
mutation_idx=mutation_idx, mutation_domain=mutation_domain, core_or_interface=core_or_interface
)
return mut_data
