def get_mutation_statistics(self, disease_name, mutation_type):
study_ids = self._get_studies_from_disease_name(disease_name)
if not study_ids:
raise DiseaseNotFoundException
gene_list_str = self._get_gene_list_str()
mutation_dict = {}
num_case = 0
for study_id in study_ids:
num_case += cbio_client.get_num_sequenced(study_id)
mutations = cbio_client.get_mutations(study_id, gene_list_str,
mutation_type)
for g, a in zip(mutations['gene_symbol'],
mutations['amino_acid_change']):
mutation_effect = self.find_mutation_effect(g, a)
if mutation_effect is None:
mutation_effect_key = 'other'
else:
mutation_effect_key = mutation_effect
try:
mutation_dict[g][0] += 1.0
mutation_dict[g][1][mutation_effect_key] += 1
except KeyError:
effect_dict = {'activate': 0.0, 'deactivate': 0.0,
'other': 0.0}
effect_dict[mutation_effect_key] += 1.0
mutation_dict[g] = [1.0, effect_dict]
# Normalize entries
for k, v in mutation_dict.iteritems():
mutation_dict[k][0] /= num_case
effect_sum = numpy.sum(mutation_dict[k][1].values())
mutation_dict[k][1]['activate'] /= effect_sum
mutation_dict[k][1]['deactivate'] /= effect_sum
mutation_dict[k][1]['other'] /= effect_sum
return mutation_dict
def get_mutation_statistics(self, disease_name, mutation_type):
study_ids = self._get_studies_from_disease_name(disease_name)
if not study_ids:
raise DiseaseNotFoundException
gene_list_str = self._get_gene_list_str()
mutation_dict = {}
num_case = 0
for study_id in study_ids:
num_case += cbio_client.get_num_sequenced(study_id)
mutations = cbio_client.get_mutations(study_id, gene_list_str,
mutation_type)
for g, a in zip(mutations['gene_symbol'],
mutations['amino_acid_change']):
mutation_effect = self.find_mutation_effect(g, a)
if mutation_effect is None:
mutation_effect_key = 'other'
else:
mutation_effect_key = mutation_effect
try:
mutation_dict[g][0] += 1.0
mutation_dict[g][1][mutation_effect_key] += 1
except KeyError:
effect_dict = {'activate': 0.0, 'deactivate': 0.0,
'other': 0.0}
effect_dict[mutation_effect_key] += 1.0
mutation_dict[g] = [1.0, effect_dict]
# Normalize entries
for k, v in mutation_dict.iteritems():
mutation_dict[k][0] /= num_case
effect_sum = numpy.sum(mutation_dict[k][1].values())
mutation_dict[k][1]['activate'] /= effect_sum
mutation_dict[k][1]['deactivate'] /= effect_sum
mutation_dict[k][1]['other'] /= effect_sum
return mutation_dict
def test_get_num_sequenced():
num_case = cbio_client.get_num_sequenced('paad_tcga')
assert (num_case > 0)
def test_get_num_sequenced():
num_case = cbio_client.get_num_sequenced('paad_tcga')
assert(num_case > 0)
