def _write_domain_structure_file(self, pdb_id, pdb_chains,
pdb_domain_defs):
"""Write a pdb file containing template domain chains (cut to domain bounaries)."""
if conf.CONFIGS["pdb_dir"] is not None:
pdb_file = structure_tools.get_pdb_file(pdb_id,
conf.CONFIGS["pdb_dir"],
"ent")
elif conf.CONFIGS["allow_internet"]:
pdb_file = structure_tools.download_pdb_file(
pdb_id, conf.CONFIGS["unique_temp_dir"])
else:
raise Exception
sp = structure_tools.StructureParser(pdb_file, pdb_chains,
pdb_domain_defs)
sp.extract()
sp.save_structure(conf.CONFIGS["unique_temp_dir"])
sp.save_sequences(conf.CONFIGS["unique_temp_dir"])
structure_file = op.join(conf.CONFIGS["unique_temp_dir"],
sp.pdb_id + "".join(sp.chain_ids) + ".pdb")
structure_seqrecords = []
for pdb_chain in pdb_chains:
chain_sequence = sp.chain_sequence_dict[pdb_chain]
seqrecord = SeqRecord(id=pdb_id + pdb_chain,
seq=Seq(chain_sequence))
structure_seqrecords.append(seqrecord)
return structure_file, structure_seqrecords
def run_pdb_mutation_pipeline(pdb_id, pdb_mutatations):
pdb_file = structure_tools.download_pdb_file(
pdb_id, conf.CONFIGS["unique_temp_dir"])
for chain_id, mutations in pdb_mutatations:
for mutation in mutations:
mutation_pdb = "{}_{}".format(chain_id, mutation)
lp = standalone_pipeline.StandalonePipeline(pdb_file,
mutations=mutation_pdb)
lp.run_all_sequences()
lp.run_all_models()
lp.run_all_mutations()
logger.info("Pass")
def test_foldx(test_data, tmpdir):
tmp_dir = str(tmpdir.mkdtemp())
pdb_file = download_pdb_file(test_data["pdb_id"], tmp_dir)
pdb_chains = (test_data["pdb_chain"], test_data["pdb_partner_chain"])
foldx_mutation = (test_data["pdb_mutation"][0] + test_data["pdb_chain"] +
test_data["pdb_mutation"][1:])
foldx = FoldX(tmp_dir)
structure_file_wt, structure_file_mut = _test_build_model(
pdb_file, foldx_mutation, foldx, test_data["stability_results_"])
_test_stability(structure_file_wt, structure_file_mut, foldx,
test_data["stability_results_"])
_test_analyse_complex(
structure_file_wt,
structure_file_mut,
pdb_chains,
foldx,
test_data["analyze_complex_results_"],
)
def run_sequence_mutation_pipeline(pdb_id_sequence, sequence_mutations):
"""
Parameters
----------
pdb_id_sequence : tuple
(pdb_id, uniprot_id)
"""
pdb_id, sequence_id = pdb_id_sequence
pdb_file = structure_tools.download_pdb_file(pdb_id, conf.CONFIGS["unique_temp_dir"])
sequence_file = elaspic_sequence.download_uniport_sequence(sequence_id, conf.CONFIGS["unique_temp_dir"])
for chain_pos in sequence_mutations[pdb_id_sequence]:
for mutation in sequence_mutations[pdb_id_sequence][chain_pos]:
mutation_sequence = "{}_{}".format(chain_pos, mutation)
lp = standalone_pipeline.StandalonePipeline(pdb_file, sequence_file, mutations=mutation_sequence)
lp.run_all_sequences()
lp.run_all_models()
lp.run_all_mutations()
logger.info("Done!\n")
logger.info("Pass")
def run_pdb_mutation_pipeline(pdb_id, pdb_mutatations, have_sequences=False, have_models=False, have_mutations=False):
pdb_file = structure_tools.download_pdb_file(pdb_id, conf.CONFIGS["unique_temp_dir"])
for chain_id in pdb_mutatations[pdb_id]:
for mutation in pdb_mutatations[pdb_id][chain_id]:
mutation_pdb = "{}_{}".format(chain_id, mutation)
lp = standalone_pipeline.StandalonePipeline(pdb_file, mutations=mutation_pdb)
if have_sequences:
assert lp.have_sequences
else:
lp.run_all_sequences()
if have_models:
assert lp.have_models
else:
lp.run_all_models()
if have_mutations:
assert lp.have_mutations
else:
lp.run_all_mutations()
logger.info("Pass")
def _write_domain_structure_file(self, pdb_id, pdb_chains, pdb_domain_defs):
"""Write a pdb file containing template domain chains (cut to domain bounaries)."""
pdb_file = structure_tools.get_pdb_file(pdb_id, conf.CONFIGS["pdb_dir"], "ent")
if not op.isfile(pdb_file) and conf.CONFIGS["allow_internet"]:
pdb_file = structure_tools.download_pdb_file(pdb_id, conf.CONFIGS["pdb_dir"])
sp = structure_tools.StructureParser(pdb_file, pdb_chains, pdb_domain_defs)
sp.extract()
sp.save_structure(conf.CONFIGS["unique_temp_dir"])
sp.save_sequences(conf.CONFIGS["unique_temp_dir"])
structure_file = op.join(conf.CONFIGS["unique_temp_dir"], sp.pdb_id + "".join(sp.chain_ids) + ".pdb")
structure_seqrecords = []
for pdb_chain in pdb_chains:
chain_sequence = sp.chain_sequence_dict[pdb_chain]
seqrecord = SeqRecord(id=pdb_id + pdb_chain, seq=Seq(chain_sequence))
structure_seqrecords.append(seqrecord)
return structure_file, structure_seqrecords
def get_structure(pdb_id, input_folder, output_folder, use_remote=True):
"""Move PDB structure to the local working directory."""
input_file = op.join(input_folder, pdb_id + ".pdb")
output_file = op.join(output_folder, pdb_id + ".pdb")
# If the PDB already exists, do nothing...
if op.isfile(output_file):
logger.debug("Structure file {} already exists!".format(output_file))
return output_file
# Look for PDB file in the same folder
if not op.isfile(input_file):
if use_remote:
input_file = structure_tools.download_pdb_file(pdb_id, input_folder)
else:
raise Exception("No PDB input file found!")
logger.info("Copying {} to {}...".format(input_file, output_file))
shutil.copy(input_file, output_file)
return output_file
def run_sequence_mutation_pipeline(pdb_id_sequence, sequence_mutations):
"""
Parameters
----------
pdb_id_sequence : tuple
(pdb_id, uniprot_id)
"""
pdb_id, sequence_id = pdb_id_sequence
pdb_file = structure_tools.download_pdb_file(
pdb_id, conf.CONFIGS["unique_temp_dir"])
sequence_file = elaspic_sequence.download_uniport_sequence(
sequence_id, conf.CONFIGS["unique_temp_dir"])
for chain_pos, mutations in sequence_mutations:
for mutation in mutations:
mutation_sequence = "{}_{}".format(chain_pos, mutation)
lp = standalone_pipeline.StandalonePipeline(
pdb_file, sequence_file, mutations=mutation_sequence)
lp.run_all_sequences()
lp.run_all_models()
lp.run_all_mutations()
logger.info("Pass")
def get_structure(pdb_id, input_folder, output_folder, use_remote=True):
"""Move PDB structure to the local working directory."""
input_file = op.join(input_folder, pdb_id + ".pdb")
output_file = op.join(output_folder, pdb_id + ".pdb")
# If the PDB already exists, do nothing...
if op.isfile(output_file):
logger.debug("Structure file {} already exists!".format(output_file))
return output_file
# Look for PDB file in the same folder
if not op.isfile(input_file):
if use_remote:
input_file = structure_tools.download_pdb_file(
pdb_id, input_folder)
else:
raise Exception("No PDB input file found!")
logger.info("Copying {} to {}...".format(input_file, output_file))
shutil.copy(input_file, output_file)
return output_file
