def write_output(self, dir_out):
write_string_list_to_file(self.nsv_list,
dir_out + '/' + config.NSV_LIST_FILE)
# Contains traits without a mapping in Gary's xls
with utilities.open_file(
dir_out + '/' + config.UNMAPPED_TRAITS_FILE_NAME, 'wt') as fdw:
fdw.write('Trait\tCount\n')
for trait_list in self.unmapped_traits:
fdw.write(
str(trait_list) + '\t' +
str(self.unmapped_traits[trait_list]) + '\n')
# Contains urls provided by Gary which are not yet included within EFO
with utilities.open_file(
dir_out + '/' + config.UNAVAILABLE_EFO_FILE_NAME, 'wt') as fdw:
fdw.write('Trait\tCount\n')
for clinvar_name in self.unavailable_efo:
fdw.write(clinvar_name + "\n")
with utilities.open_file(
dir_out + '/' + config.EVIDENCE_STRINGS_FILE_NAME,
'wt') as fdw:
for evidence_string in self.evidence_string_list:
fdw.write(json.dumps(evidence_string) + '\n')
self.write_zooma_file(dir_out)
def test_existing(self):
test_file_b = os.path.join(self.test_dir_b, "test.txt")
with utilities.open_file(test_file_b, "wt") as f:
f.write("hello world")
utilities.copy_dir(self.test_dir_a, self.test_dir_b)
with utilities.open_file(test_file_b, "rt") as f:
contents = f.read()
self.assertEqual(contents, self.test_string)
def test_existing(self):
test_file_b = os.path.join(self.test_dir_b, "test.txt")
with utilities.open_file(test_file_b, "wt") as f:
f.write("hello world")
utilities.copy_dir(self.test_dir_a, self.test_dir_b)
with utilities.open_file(test_file_b, "rt") as f:
contents = f.read()
self.assertEqual(contents, self.test_string)
def test_nonexisting(self):
utilities.copy_dir(self.test_dir_a, self.test_dir_c)
with utilities.open_file(os.path.join(self.test_dir_c, "test.txt"),
"rt") as f:
contents = f.read()
self.assertEqual(contents, self.test_string)
shutil.rmtree(self.test_dir_c)
def process_consequence_type_file_tsv(snp_2_gene_filepath):
consequence_type_dict = defaultdict(list)
one_rs_multiple_genes = set()
with utilities.open_file(snp_2_gene_filepath, "rt") as snp_2_gene_file:
for line in snp_2_gene_file:
line = line.rstrip()
line_list = line.split("\t")
if len(line_list) < 6:
logger.warning(
'Skip invalid line in snp_2_gene file: {}'.format(line))
continue
variant_id = line_list[0]
ensembl_gene_id = line_list[2]
so_term = line_list[4]
if ensembl_gene_id == 'NA':
logger.warning(
'Skip line with missing gene ID: {}'.format(line))
continue
process_gene(consequence_type_dict, variant_id, ensembl_gene_id,
so_term)
return consequence_type_dict, one_rs_multiple_genes
def get_test_record():
test_clinvar_record_filepath = os.path.join(os.path.dirname(__file__), 'resources',
'test_clinvar_record.json')
with utilities.open_file(test_clinvar_record_filepath, "rt") as f:
test_record_dict = json.load(f)
test_record = clinvar.ClinvarRecord(test_record_dict)
return test_record
def load_efo_mapping(efo_mapping_file):
trait_2_efo = defaultdict(list)
n_efo_mappings = 0
with utilities.open_file(efo_mapping_file, "rt") as f:
for line in f:
line = line.rstrip()
if line.startswith("#") or not line:
continue
line_list = line.split("\t")
clinvar_name = line_list[0].lower()
if len(line_list) > 1:
ontology_id_list = line_list[1].split("|")
ontology_label_list = line_list[2].split("|") if len(
line_list) > 2 else [None] * len(ontology_id_list)
for ontology_id, ontology_label in zip(ontology_id_list,
ontology_label_list):
trait_2_efo[clinvar_name].append(
(ontology_id, ontology_label))
n_efo_mappings += 1
else:
raise ValueError(
'No mapping provided for trait: {}'.format(clinvar_name))
logger.info('{} EFO mappings loaded'.format(n_efo_mappings))
return trait_2_efo
def load_efo_mapping(efo_mapping_file):
trait_2_efo = defaultdict(list)
unavailable_efo = set()
n_efo_mappings = 0
with utilities.open_file(efo_mapping_file, "rt") as f:
for line in f:
if line.startswith("#"):
continue
line_list = line.rstrip().split("\t")
clinvar_name = line_list[0].lower()
if len(line_list) > 1:
ontology_id_list = line_list[1].split("|")
ontology_label_list = line_list[2].split("|") if len(
line_list) > 2 else [None] * len(ontology_id_list)
for ontology_id, ontology_label in zip(ontology_id_list,
ontology_label_list):
trait_2_efo[clinvar_name].append(
(ontology_id, ontology_label))
n_efo_mappings += 1
else:
unavailable_efo.add(clinvar_name)
print(str(n_efo_mappings) + ' EFO mappings loaded')
print(
str(len(unavailable_efo)) +
' urls without an actual valid EFO mapping')
return trait_2_efo, unavailable_efo
def get_terms_from_file(terms_file_path):
if terms_file_path is not None:
print('Loading list of terms...')
with utilities.open_file(terms_file_path, 'rt') as terms_file:
terms_list = [line.rstrip() for line in terms_file]
print(str(len(terms_file_path)) + ' terms found at ' + terms_file_path)
else:
terms_list = []
return terms_list
def setUp(self):
self.test_dir_a = os.path.join(os.path.dirname(__file__), "resources", "test_tmp_a")
self.test_dir_b = os.path.join(os.path.dirname(__file__), "resources", "test_tmp_b")
self.test_dir_c = os.path.join(os.path.dirname(__file__), "resources", "test_tmp_c")
os.makedirs(self.test_dir_a)
os.makedirs(self.test_dir_b)
self.test_file_a = os.path.join(self.test_dir_a, "test.txt")
self.test_string = "this is a test string"
with utilities.open_file(self.test_file_a, "wt") as f:
f.write(self.test_string)
def get_terms_from_file(terms_file_path):
if terms_file_path is not None:
print('Loading list of terms...')
with utilities.open_file(terms_file_path, 'rt') as terms_file:
terms_list = [line.rstrip() for line in terms_file]
print(str(len(terms_file_path)) + ' terms found at ' + terms_file_path)
else:
terms_list = []
return terms_list
def write_zooma_file(self, dir_out):
"""Write zooma records to zooma file"""
with utilities.open_file(os.path.join(dir_out, config.ZOOMA_FILE_NAME), "wt") as zooma_fh:
zooma_fh.write("STUDY\tBIOENTITY\tPROPERTY_TYPE\tPROPERTY_VALUE\tSEMANTIC_TAG\tANNOTATOR\tANNOTATION_DATE\n")
date = strftime("%d/%m/%y %H:%M", gmtime())
for evidence_record in self.evidence_list:
self.write_zooma_record_to_zooma_file(evidence_record, zooma_fh, date)
for trait_name, ontology_tuple_list in self.trait_mappings.items():
self.write_extra_trait_to_zooma_file(ontology_tuple_list, trait_name, date, zooma_fh)
def write_output(self, dir_out):
write_string_list_to_file(self.nsv_list,
dir_out + '/' + config.NSV_LIST_FILE)
# Contains traits without a mapping in Gary's xls
with utilities.open_file(
dir_out + '/' + config.UNMAPPED_TRAITS_FILE_NAME, 'wt') as fdw:
fdw.write('Trait\tCount\n')
for trait_list in self.unmapped_traits:
fdw.write(
str(trait_list) + '\t' +
str(self.unmapped_traits[trait_list]) + '\n')
with utilities.open_file(
dir_out + '/' + config.EVIDENCE_STRINGS_FILE_NAME,
'wt') as fdw:
for evidence_string in self.evidence_string_list:
fdw.write(json.dumps(evidence_string) + '\n')
self.write_zooma_file(dir_out)
def write_output(self, dir_out):
write_string_list_to_file(self.nsv_list, dir_out + '/' + config.NSV_LIST_FILE)
# Contains traits without a mapping in Gary's xls
with utilities.open_file(dir_out + '/' + config.UNMAPPED_TRAITS_FILE_NAME, 'wt') as fdw:
fdw.write('Trait\tCount\n')
for trait_list in self.unmapped_traits:
fdw.write(str(trait_list) + '\t' +
str(self.unmapped_traits[trait_list]) + '\n')
# Contains urls provided by Gary which are not yet included within EFO
with utilities.open_file(dir_out + '/' + config.UNAVAILABLE_EFO_FILE_NAME, 'wt') as fdw:
fdw.write('Trait\tCount\n')
for clinvar_name in self.unavailable_efo:
fdw.write(clinvar_name + "\n")
with utilities.open_file(dir_out + '/' + config.EVIDENCE_STRINGS_FILE_NAME, 'wt') as fdw:
for evidence_string in self.evidence_string_list:
fdw.write(json.dumps(evidence_string) + '\n')
self.write_zooma_file(dir_out)
def setUp(self):
self.test_dir_a = os.path.join(os.path.dirname(__file__), "resources",
"test_tmp_a")
self.test_dir_b = os.path.join(os.path.dirname(__file__), "resources",
"test_tmp_b")
# self.test_dir_c = os.path.join(os.path.dirname(__file__), "resources", "test_tmp_c")
os.makedirs(self.test_dir_a)
os.makedirs(self.test_dir_b)
self.test_file_a = os.path.join(self.test_dir_a, "test.txt")
self.test_string = "this is a test string"
with utilities.open_file(self.test_file_a, "wt") as f:
f.write(self.test_string)
def write_zooma_file(self, dir_out):
"""Write zooma records to zooma file"""
with utilities.open_file(os.path.join(dir_out, config.ZOOMA_FILE_NAME),
"wt") as zooma_fh:
zooma_fh.write(
"STUDY\tBIOENTITY\tPROPERTY_TYPE\tPROPERTY_VALUE\tSEMANTIC_TAG\tANNOTATOR\tANNOTATION_DATE\n"
)
date = strftime("%d/%m/%y %H:%M", gmtime())
for evidence_record in self.evidence_list:
self.write_zooma_record_to_zooma_file(evidence_record,
zooma_fh, date)
for trait_name, ontology_tuple_list in self.trait_mappings.items():
self.write_extra_trait_to_zooma_file(ontology_tuple_list,
trait_name, date,
zooma_fh)
def process_consequence_type_file_tsv(snp_2_gene_filepath):
consequence_type_dict = defaultdict(list)
one_rs_multiple_genes = set()
with utilities.open_file(snp_2_gene_filepath, "rt") as snp_2_gene_file:
for line in snp_2_gene_file:
line = line.rstrip()
line_list = line.split("\t")
if len(line_list) < 6:
continue
variant_id = line_list[0]
ensembl_gene_id = line_list[2]
so_term = line_list[4]
process_gene(consequence_type_dict, variant_id, ensembl_gene_id, so_term)
return consequence_type_dict, one_rs_multiple_genes
def process_consequence_type_file_tsv(snp_2_gene_filepath):
consequence_type_dict = defaultdict(list)
one_rs_multiple_genes = set()
with utilities.open_file(snp_2_gene_filepath, "rt") as snp_2_gene_file:
for line in snp_2_gene_file:
line = line.rstrip()
line_list = line.split("\t")
if len(line_list) < 6:
continue
variant_id = line_list[0]
ensembl_gene_id = line_list[2]
so_term = line_list[4]
process_gene(consequence_type_dict, variant_id, ensembl_gene_id,
so_term)
return consequence_type_dict, one_rs_multiple_genes
def load_efo_mapping(efo_mapping_file):
trait_2_efo = defaultdict(list)
unavailable_efo = set()
n_efo_mappings = 0
with utilities.open_file(efo_mapping_file, "rt") as f:
for line in f:
if line.startswith("#"):
continue
line_list = line.rstrip().split("\t")
clinvar_name = line_list[0].lower()
if len(line_list) > 1:
ontology_id_list = line_list[1].split("|")
ontology_label_list = line_list[2].split("|") if len(line_list) > 2 else [None] * len(ontology_id_list)
for ontology_id, ontology_label in zip(ontology_id_list, ontology_label_list):
trait_2_efo[clinvar_name].append((ontology_id, ontology_label))
n_efo_mappings += 1
else:
unavailable_efo.add(clinvar_name)
print(str(n_efo_mappings) + ' EFO mappings loaded')
print(str(len(unavailable_efo)) + ' urls without an actual valid EFO mapping')
return trait_2_efo, unavailable_efo
def __iter__(self):
with utilities.open_file(self.json_file, "rt") as f:
for line in f:
yield json.loads(line.rstrip())
def clinvar_to_evidence_strings(allowed_clinical_significance, mappings,
json_file, ot_schema, output_evidence_strings):
report = Report(trait_mappings=mappings.trait_2_efo)
cell_recs = cellbase_records.CellbaseRecords(json_file=json_file)
ot_schema_contents = json.loads(open(ot_schema).read())
output_evidence_strings_file = utilities.open_file(output_evidence_strings,
'wt')
for cellbase_record in cell_recs:
report.counters["record_counter"] += 1
if report.counters["record_counter"] % 1000 == 0:
logger.info("{} records processed".format(
report.counters["record_counter"]))
n_ev_strings_per_record = 0
clinvar_record = clinvar.ClinvarRecord(cellbase_record['clinvarSet'])
for clinvar_record_measure in clinvar_record.measures:
report.counters["n_nsvs"] += (clinvar_record_measure.nsv_id
is not None)
append_nsv(report.nsv_list, clinvar_record_measure)
report.counters["n_multiple_allele_origin"] += (len(
clinvar_record.allele_origins) > 1)
traits = create_traits(clinvar_record.traits, mappings.trait_2_efo,
report)
converted_allele_origins = convert_allele_origins(
clinvar_record.allele_origins)
for consequence_type, trait, allele_origin in itertools.product(
get_consequence_types(clinvar_record_measure,
mappings.consequence_type_dict),
traits, converted_allele_origins):
if skip_record(clinvar_record, clinvar_record_measure,
consequence_type, allele_origin,
allowed_clinical_significance, report):
continue
if allele_origin == 'germline':
evidence_string = evidence_strings.CTTVGeneticsEvidenceString(
clinvar_record, clinvar_record_measure, report, trait,
consequence_type)
elif allele_origin == 'somatic':
evidence_string = evidence_strings.CTTVSomaticEvidenceString(
clinvar_record, clinvar_record_measure, report, trait,
consequence_type)
else:
raise AssertionError(
'Unknown allele_origin present in the data: {}'.format(
allele_origin))
# Validate and immediately output the evidence string (not keeping everything in memory)
validate_evidence_string(evidence_string, clinvar_record,
trait,
consequence_type.ensembl_gene_id,
ot_schema_contents)
output_evidence_strings_file.write(
json.dumps(evidence_string) + '\n')
report.evidence_string_count += 1
report.evidence_list.append([
clinvar_record.accession, clinvar_record_measure.rs_id,
trait.clinvar_name, trait.ontology_id
])
report.counters["n_valid_rs_and_nsv"] += (
clinvar_record_measure.nsv_id is not None)
report.traits.add(trait.ontology_id)
report.remove_trait_mapping(trait.clinvar_name)
report.ensembl_gene_id_uris.add(
evidence_strings.get_ensembl_gene_id_uri(
consequence_type.ensembl_gene_id))
n_ev_strings_per_record += 1
if n_ev_strings_per_record > 0:
report.counters["n_processed_clinvar_records"] += 1
if n_ev_strings_per_record > 1:
report.counters["n_multiple_evidence_strings"] += 1
output_evidence_strings_file.close()
return report
def write_string_list_to_file(string_list, filename):
with utilities.open_file(filename, 'wt') as out_file:
out_file.write('\n'.join(string_list))
def write_string_list_to_file(string_list, filename):
with utilities.open_file(filename, 'wt') as out_file:
out_file.write('\n'.join(string_list))
class CTTVGeneticsEvidenceString(CTTVEvidenceString):
"""
Class for genetics evidence string specifically.
Holds information required for Open Target's evidence strings for genetic information.
"""
with utilities.open_file(
utilities.get_resource_file(__package__,
config.GEN_EV_STRING_JSON),
"rt") as gen_json_file:
base_json = json.load(gen_json_file)
def __init__(self, clinvar_record, clinvar_record_measure, report, trait,
consequence_type):
a_dictionary = copy.deepcopy(self.base_json)
ref_list = list(
set(clinvar_record.trait_refs_list[trait.trait_counter] +
clinvar_record.observed_refs_list +
clinvar_record_measure.refs_list))
ref_list.sort()
super().__init__(a_dictionary, clinvar_record, ref_list,
consequence_type.ensembl_gene_id, report, trait)
variant_type = get_cttv_variant_type(clinvar_record_measure)
self.add_unique_association_field('alleleOrigin', 'germline')
if clinvar_record_measure.rs_id:
self.set_variant(
'http://identifiers.org/dbsnp/' + clinvar_record_measure.rs_id,
variant_type)
self.add_unique_association_field('variant_id',
clinvar_record_measure.rs_id)
elif clinvar_record_measure.nsv_id:
self.set_variant(
'http://identifiers.org/dbsnp/' +
clinvar_record_measure.nsv_id, variant_type)
self.add_unique_association_field('variant_id',
clinvar_record_measure.nsv_id)
else:
self.set_variant(
'http://www.ncbi.nlm.nih.gov/clinvar/' +
clinvar_record.accession, variant_type)
self.add_unique_association_field('variant_id',
clinvar_record.accession)
self.date = clinvar_record.date
self.db_xref_url = 'http://identifiers.org/clinvar.record/' + clinvar_record.accession
self.url = 'http://www.ncbi.nlm.nih.gov/clinvar/' + clinvar_record.accession
self.association = clinvar_record.clinical_significance not in \
('non-pathogenic', 'probable-non-pathogenic', 'likely benign', 'benign')
self.gene_2_var_ev_codes = [
'http://identifiers.org/eco/cttv_mapping_pipeline'
]
most_severe_so_term = consequence_type.so_term
if most_severe_so_term.accession is None:
self.gene_2_var_func_consequence = 'http://targetvalidation.org/sequence/' + \
most_severe_so_term.so_name
else:
self.gene_2_var_func_consequence = 'http://purl.obolibrary.org/obo/' + \
most_severe_so_term.accession.replace(':', '_')
if len(ref_list) > 0:
self.set_var_2_disease_literature(ref_list)
# Arbitrarily select only one reference among all
self.unique_reference = ref_list[0]
if clinvar_record.clinical_significance:
self.clinical_significance = clinvar_record.clinical_significance
@property
def db_xref_url(self):
if self['evidence']['gene2variant']['provenance_type']['database']['dbxref']['url'] \
== self['evidence']['variant2disease']['provenance_type']['database']['dbxref']['url']:
return \
self['evidence']['variant2disease']['provenance_type']['database']['dbxref']['url']
else:
raise Exception("db_xref_url attributes different")
@db_xref_url.setter
def db_xref_url(self, url):
self['evidence']['gene2variant']['provenance_type']['database'][
'dbxref']['url'] = url
self['evidence']['variant2disease']['provenance_type']['database'][
'dbxref']['url'] = url
@property
def url(self):
if self['evidence']['gene2variant']['urls'][0]['url'] \
== self['evidence']['variant2disease']['urls'][0]['url']:
return self['evidence']['gene2variant']['urls'][0]['url']
else:
raise Exception("url attributes different")
@url.setter
def url(self, url):
self['evidence']['gene2variant']['urls'][0]['url'] = url
self['evidence']['variant2disease']['urls'][0]['url'] = url
@property
def gene_2_var_ev_codes(self):
return self['evidence']['gene2variant']['evidence_codes']
@gene_2_var_ev_codes.setter
def gene_2_var_ev_codes(self, gene_2_var_ev_codes):
self['evidence']['gene2variant'][
'evidence_codes'] = gene_2_var_ev_codes
@property
def gene_2_var_func_consequence(self):
return self['evidence']['gene2variant']['functional_consequence']
@gene_2_var_func_consequence.setter
def gene_2_var_func_consequence(self, so_term):
self['evidence']['gene2variant']['functional_consequence'] = so_term
def set_var_2_disease_literature(self, ref_list):
self['evidence']['variant2disease']['provenance_type']['literature'] = \
{'references': [{'lit_id': reference} for reference in ref_list]}
@property
def association(self):
if self['evidence']['gene2variant']['is_associated'] \
== self['evidence']['variant2disease']['is_associated']:
return self['evidence']['gene2variant']['is_associated']
else:
raise Exception("association attributes different")
@association.setter
def association(self, is_associated):
self['evidence']['gene2variant']['is_associated'] = is_associated
self['evidence']['variant2disease']['is_associated'] = is_associated
def _clear_variant(self):
self['variant']['id'] = []
self['variant']['type'] = []
def set_variant(self, var_id, var_type):
self['variant']['id'] = var_id
self['variant']['type'] = var_type
@property
def unique_reference(self):
return self['evidence']['variant2disease'][
'unique_experiment_reference']
@unique_reference.setter
def unique_reference(self, reference):
self['evidence']['variant2disease'][
'unique_experiment_reference'] = reference
@property
def date(self):
if self['evidence']['gene2variant']['date_asserted'] == \
self['evidence']['variant2disease']['date_asserted']:
return self['evidence']['gene2variant']['date_asserted']
else:
raise Exception("date attributes have different values")
@date.setter
def date(self, date_string):
self['evidence']['gene2variant']['date_asserted'] = date_string
self['evidence']['variant2disease']['date_asserted'] = date_string
@property
def clinical_significance(self):
return self['evidence']['variant2disease']['clinical_significance']
@clinical_significance.setter
def clinical_significance(self, clinical_significance):
self['evidence']['variant2disease'][
'clinical_significance'] = clinical_significance
def test_nonexisting(self):
utilities.copy_dir(self.test_dir_a, self.test_dir_c)
with utilities.open_file(os.path.join(self.test_dir_c, "test.txt"), "rt") as f:
contents = f.read()
self.assertEqual(contents, self.test_string)
shutil.rmtree(self.test_dir_c)
class CTTVSomaticEvidenceString(CTTVEvidenceString):
"""
Class for somatic evidence string specifically.
Holds information required for Open Target's evidence strings for somatic information.
"""
with utilities.open_file(
utilities.get_resource_file(__package__,
config.SOM_EV_STRING_JSON),
"rt") as som_json_file:
base_json = json.load(som_json_file)
def __init__(self, clinvar_record, clinvar_record_measure, report, trait,
consequence_type):
a_dictionary = copy.deepcopy(self.base_json)
ref_list = list(
set(clinvar_record.trait_refs_list[trait.trait_counter] +
clinvar_record.observed_refs_list +
clinvar_record_measure.refs_list))
ref_list.sort()
super().__init__(a_dictionary, clinvar_record, ref_list,
consequence_type.ensembl_gene_id, report, trait)
self.add_unique_association_field('alleleOrigin', 'somatic')
if clinvar_record_measure.rs_id:
self.add_unique_association_field('variant_id',
clinvar_record_measure.rs_id)
elif clinvar_record_measure.nsv_id:
self.add_unique_association_field('variant_id',
clinvar_record_measure.nsv_id)
else:
self.add_unique_association_field('variant_id',
clinvar_record.accession)
self.date = clinvar_record.date
self.last_evaluated_date = clinvar_record.last_evaluated_date
self.db_xref_url = 'http://identifiers.org/clinvar.record/' + clinvar_record.accession
self.url = 'http://www.ncbi.nlm.nih.gov/clinvar/' + clinvar_record.accession
# See https://github.com/opentargets/platform/issues/1139#issuecomment-682592678
self.association = True
self.set_known_mutations(consequence_type.so_term)
if len(ref_list) > 0:
self.evidence_literature = ref_list
if clinvar_record.clinical_significance:
self.clinical_significance = process_clinical_significance(
clinvar_record.clinical_significance)
# Populate star rating and review status
star_rating, review_status = clinvar_record.score
self.clinvar_rating = (star_rating, review_status)
# Populate mode of inheritance (if present)
self.mode_of_inheritance = clinvar_record.mode_of_inheritance
@property
def db_xref_url(self):
return self['evidence']['provenance_type']['database']['dbxref']['url']
@db_xref_url.setter
def db_xref_url(self, url):
self['evidence']['provenance_type']['database']['dbxref']['url'] = url
@property
def url(self):
return self['evidence']['urls'][0]['url']
@url.setter
def url(self, url):
self['evidence']['urls'][0]['url'] = url
@property
def evidence_literature(self):
return self['evidence']['provenance_type']['literature']['references']
@evidence_literature.setter
def evidence_literature(self, ref_list):
self['evidence']['provenance_type']['literature'] = \
{'references': [{'lit_id': reference} for reference in ref_list]}
@property
def association(self):
return self['evidence']['is_associated']
@association.setter
def association(self, is_associated):
self['evidence']['is_associated'] = is_associated
@property
def date(self):
return self['evidence']['date_asserted']
@date.setter
def date(self, date_string):
self['evidence']['date_asserted'] = date_string
@property
def last_evaluated_date(self):
return self['evidence']['last_evaluated_date']
@last_evaluated_date.setter
def last_evaluated_date(self, clinvar_last_evaluated_date):
if clinvar_last_evaluated_date:
self['evidence'][
'last_evaluated_date'] = clinvar_last_evaluated_date
def _clear_known_mutations(self):
self['evidence']['known_mutations'] = []
def add_known_mutation(self, new_functional_consequence, so_name):
new_known_mutation = \
{'functional_consequence': new_functional_consequence, 'preferred_name': so_name}
self['evidence']['known_mutations'].append(new_known_mutation)
def set_known_mutations(self, so_term):
if so_term.accession:
new_functional_consequence = \
"http://purl.obolibrary.org/obo/" + so_term.accession.replace(':', '_')
else:
new_functional_consequence = \
'http://targetvalidation.org/sequence/' + so_term.so_name
self.add_known_mutation(new_functional_consequence, so_term.so_name)
@property
def clinical_significance(self):
return self['evidence']['clinical_significance']
@clinical_significance.setter
def clinical_significance(self, clinical_significance):
self['evidence']['clinical_significance'] = clinical_significance
@property
def clinvar_rating(self):
return self['evidence']['clinvar_rating']
@clinvar_rating.setter
def clinvar_rating(self, clinvar_rating_data):
star_rating, review_status = clinvar_rating_data
self['evidence']['clinvar_rating'] = {
'star_rating': star_rating,
'review_status': review_status,
}
@property
def mode_of_inheritance(self):
return self['evidence'].get('mode_of_inheritance')
@mode_of_inheritance.setter
def mode_of_inheritance(self, mode_of_inheritance):
if mode_of_inheritance:
self['evidence']['mode_of_inheritance'] = mode_of_inheritance
class CTTVSomaticEvidenceString(CTTVEvidenceString):
"""
Class for somatic evidence string specifically.
Holds information required for Open Target's evidence strings for somatic information.
"""
with utilities.open_file(
utilities.get_resource_file(__package__,
config.SOM_EV_STRING_JSON),
"rt") as som_json_file:
base_json = json.load(som_json_file)
def __init__(self, clinvar_record, clinvar_record_measure, report, trait,
consequence_type):
a_dictionary = copy.deepcopy(self.base_json)
ref_list = list(
set(clinvar_record.trait_refs_list[trait.trait_counter] +
clinvar_record.observed_refs_list +
clinvar_record_measure.refs_list))
ref_list.sort()
super().__init__(a_dictionary, clinvar_record, ref_list,
consequence_type.ensembl_gene_id, report, trait)
self.add_unique_association_field('alleleOrigin', 'somatic')
if clinvar_record_measure.rs_id:
self.add_unique_association_field('variant_id',
clinvar_record_measure.rs_id)
elif clinvar_record_measure.nsv_id:
self.add_unique_association_field('variant_id',
clinvar_record_measure.nsv_id)
else:
self.add_unique_association_field('variant_id',
clinvar_record.accession)
self.date = clinvar_record.date
self.db_xref_url = 'http://identifiers.org/clinvar.record/' + clinvar_record.accession
self.url = 'http://www.ncbi.nlm.nih.gov/clinvar/' + clinvar_record.accession
self.association = clinvar_record.clinical_significance not in \
('non-pathogenic', 'probable-non-pathogenic', 'likely benign', 'benign')
self.set_known_mutations(consequence_type.so_term)
if len(ref_list) > 0:
self.evidence_literature = ref_list
if clinvar_record.clinical_significance:
self.clinical_significance = clinvar_record.clinical_significance
@property
def db_xref_url(self):
return self['evidence']['provenance_type']['database']['dbxref']['url']
@db_xref_url.setter
def db_xref_url(self, url):
self['evidence']['provenance_type']['database']['dbxref']['url'] = url
@property
def url(self):
return self['evidence']['urls'][0]['url']
@url.setter
def url(self, url):
self['evidence']['urls'][0]['url'] = url
@property
def evidence_literature(self):
return self['evidence']['provenance_type']['literature']['references']
@evidence_literature.setter
def evidence_literature(self, ref_list):
self['evidence']['provenance_type']['literature'] = \
{'references': [{'lit_id': reference} for reference in ref_list]}
@property
def association(self):
return self['evidence']['is_associated']
@association.setter
def association(self, is_associated):
self['evidence']['is_associated'] = is_associated
@property
def date(self):
return self['evidence']['date_asserted']
@date.setter
def date(self, date_string):
self['evidence']['date_asserted'] = date_string
def _clear_known_mutations(self):
self['evidence']['known_mutations'] = []
def add_known_mutation(self, new_functional_consequence, so_name):
new_known_mutation = \
{'functional_consequence': new_functional_consequence, 'preferred_name': so_name}
self['evidence']['known_mutations'].append(new_known_mutation)
def set_known_mutations(self, so_term):
if so_term.accession:
new_functional_consequence = \
"http://purl.obolibrary.org/obo/" + so_term.accession.replace(':', '_')
else:
new_functional_consequence = \
'http://targetvalidation.org/sequence/' + so_term.so_name
self.add_known_mutation(new_functional_consequence, so_term.so_name)
@property
def clinical_significance(self):
return self['evidence']['clinical_significance']
@clinical_significance.setter
def clinical_significance(self, clinical_significance):
self['evidence']['clinical_significance'] = clinical_significance
def __iter__(self):
with utilities.open_file(self.json_file, "rt") as f:
for line in f:
yield json.loads(line.rstrip())
