class LookUp(object):
def __init__(self):
self.blw = BioLinkWrapper(Config().get_biolink_api_endpoint())
self.mg = get_client('gene')
self.input_object = ''
self.meta = {
'data_type': 'disease',
'input_type': {
'complexity': 'single',
'id_type': ['MONDO', 'DO', 'OMIM'],
},
'output_type': {
'complexity': 'set',
'id_type': 'HGNC'
},
'taxon': 'human',
'limit': None,
'source': 'Monarch Biolink',
'predicate': 'blm:gene associated with condition'
}
def metadata(self):
print("""Mod O DiseaseGeneLookup metadata:""")
pprint(self.meta)
#
# def load_input_object(self, input_object):
# input_object = self.blw.get_obj(obj_id=input_object['input'])
# self.input_object = {
# 'id': input_object['id'],
# 'label': input_object['label'],
# 'description': input_object['description'],
# }
#
# def get_input_object_id(self):
# return self.input_object['id']
def echo_input_object(self, output=None):
if output:
dump(self.input_object, output, indent=4, separators=(',', ': '))
else:
dump(self.input_object, stdout, indent=4, separators=(',', ': '))
def disease_geneset_lookup(self, name, id):
input_disease_label = name
input_disease_id = id
input_gene_set = self.blw.disease2genes(input_disease_id)
input_gene_set = [self.blw.parse_association(input_disease_id, input_disease_label, x) for x in
input_gene_set['associations']]
for input_gene in input_gene_set:
igene_mg = self.mg.query(input_gene['hit_id'].replace('HGNC', 'hgnc'), species='human', entrezonly=True,
fields='entrez,HGNC,symbol')
input_gene.update({'input_ncbi': 'NCBIGene:{}'.format(igene_mg['hits'][0]['_id'])})
input_genes_df = pd.DataFrame(data=input_gene_set)
# # group duplicate ids and gather sources
input_genes_df['sources'] = input_genes_df['sources'].str.join(', ')
input_genes_df = input_genes_df.groupby(
['input_id', 'input_symbol', 'hit_id', 'hit_symbol', 'relation'])['sources'].apply(', '.join).reset_index()
return input_genes_df
class GeneInteractions(object):
def __init__(self):
self.blw = BioLinkWrapper()
self.meta = {
'input_type': {
'complexity': 'set',
'id_type': 'HGNC',
'data_type': 'gene',
},
'output_type': {
'complexity': 'set',
'id_type': 'HGNC',
'data_type': 'gene',
},
'source': 'Monarch Biolink',
'predicate': ['blm:interacts with']
}
def metadata(self):
print("""Mod1E Interaction Network metadata:""")
pprint(self.meta)
@staticmethod
def load_gene_set(input_gene_set):
annotated_gene_set = []
for gene in input_gene_set.get_input_curie_set():
annotated_gene_set.append({
'input_id': gene['hit_id'],
'sim_input_curie': gene['hit_id'],
'input_symbol': gene['hit_symbol']
})
return annotated_gene_set
def get_interactions(self, annotated_gene_set):
results = []
for gene in annotated_gene_set:
interactions = self.blw.gene_interactions(gene_curie=gene['sim_input_curie'])
for assoc in interactions['associations']:
interaction = \
self.blw.parse_association(
input_id=gene['sim_input_curie'],
input_label=gene['input_symbol'],
association=assoc
)
results.append({
'input_id': interaction['input_id'],
'input_symbol': interaction['input_symbol'],
'hit_symbol': interaction['hit_symbol'],
'hit_id': interaction['hit_id'],
'score': 0,
})
return results
class LookUp(object):
def __init__(self):
self.blw = BioLinkWrapper()
self.mg = MyGeneInfo()
self.input_object = ''
self.meta = {
'data_type': 'disease',
'input_type': {
'complexity': 'single',
'id_type': ['MONDO', 'DO', 'OMIM'],
},
'output_type': {
'complexity': 'set',
'id_type': 'HGNC'
},
'taxon': 'human',
'limit': None,
'source': 'Monarch Biolink',
'predicate': 'blm:gene associated with condition'
}
print("""Mod O DiseaseGeneLookup metadata:""")
pprint(self.meta)
def load_input_object(self, input_object):
input_object = self.blw.get_obj(obj_id=input_object['input'])
self.input_object = {
'id': input_object['id'],
'label': input_object['label'],
'description': input_object['description'],
}
pprint(self.input_object)
def disease_geneset_lookup(self):
input_disease_id = self.input_object['id']
input_disease_label = self.input_object['label']
input_gene_set = self.blw.disease2genes(input_disease_id)
input_gene_set = [
self.blw.parse_association(input_disease_id, input_disease_label,
x)
for x in input_gene_set['associations']
]
# for input_gene in input_gene_set:
# igene_mg = self.mg.query(input_gene['hit_id'].replace('HGNC', 'hgnc'), species='human', entrezonly=True,
# fields='entrez,HGNC,symbol')
# input_gene.update({'input_ncbi': 'NCBIGene:{}'.format(igene_mg['hits'][0]['_id'])})
input_genes_df = pd.DataFrame(data=input_gene_set)
# # group duplicate ids and gather sources
input_genes_df['sources'] = input_genes_df['sources'].str.join(', ')
input_genes_df = input_genes_df.groupby(
['input_id', 'input_symbol', 'hit_id', 'hit_symbol',
'relation'])['sources'].apply(', '.join).reset_index()
return input_genes_df
class GeneInteractions(object):
def __init__(self):
self.blw = BioLinkWrapper()
self.gene_set = []
self.input_object = ''
self.meta = {
'input_type': {
'complexity': 'set',
'id_type': 'HGNC',
'data_type': 'gene',
},
'output_type': {
'complexity': 'set',
'id_type': 'HGNC',
'data_type': 'gene',
},
'source': 'Monarch Biolink',
'predicate': ['blm:interacts with']
}
def load_input_object(self, input_object):
self.input_object = input_object
def load_gene_set(self):
for gene in self.input_object['input']:
self.gene_set.append({
'input_id': gene['hit_id'],
'sim_input_curie': gene['hit_id'],
'input_symbol': gene['hit_symbol']
})
def get_interactions(self):
results = []
for gene in self.gene_set:
interactions = self.blw.gene_interactions(
gene_curie=gene['sim_input_curie'])
for assoc in interactions['associations']:
interaction = self.blw.parse_association(
input_id=gene['sim_input_curie'],
input_label=gene['input_symbol'],
association=assoc)
results.append({
'input_id': interaction['input_id'],
'input_symbol': interaction['input_symbol'],
'hit_symbol': interaction['hit_symbol'],
'hit_id': interaction['hit_id'],
'score': 0,
})
return results
def __init__(self):
self.blw = BioLinkWrapper(Config().getBiolinkApiEndpoint())
self.meta = {
'input_type': {
'complexity': 'set',
'id_type': 'HGNC',
'data_type': 'gene',
},
'output_type': {
'complexity': 'set',
'id_type': 'HGNC',
'data_type': 'gene',
},
'source': 'Monarch Biolink',
'predicate': ['blm:interacts with']
}
def __init__(self):
self.blw = BioLinkWrapper()
self.gene_set = []
self.input_object = ''
self.meta = {
'input_type': {
'complexity': 'set',
'id_type': 'HGNC',
'data_type': 'gene',
},
'output_type': {
'complexity': 'set',
'id_type': 'HGNC',
'data_type': 'gene',
},
'source': 'Monarch Biolink',
'predicate': ['blm:interacts with']
}
def __init__(self):
self.blw = BioLinkWrapper(Config().get_biolink_api_endpoint())
self.mg = get_client('gene')
self.input_object = ''
self.meta = {
'data_type': 'disease',
'input_type': {
'complexity': 'single',
'id_type': ['MONDO', 'DO', 'OMIM'],
},
'output_type': {
'complexity': 'set',
'id_type': 'HGNC'
},
'taxon': 'human',
'limit': None,
'source': 'Monarch Biolink',
'predicate': 'blm:gene associated with condition'
}
def __init__(self):
self.blw = BioLinkWrapper()
self.mg = MyGeneInfo()
self.input_object = ''
self.meta = {
'data_type': 'disease',
'input_type': {
'complexity': 'single',
'id_type': ['MONDO', 'DO', 'OMIM'],
},
'output_type': {
'complexity': 'set',
'id_type': 'HGNC'
},
'taxon': 'human',
'limit': None,
'source': 'Monarch Biolink',
'predicate': 'blm:gene associated with condition'
}
def __init__(self):
self.blw = BioLinkWrapper()
self.gene_set = []
self.input_object = ''
self.meta = {
'input_type': {
'complexity': 'set',
'id_type': 'HGNC',
'data_type': 'gene',
},
'output_type': {
'complexity': 'set',
'id_type': 'HGNC',
'data_type': 'gene',
},
'source': 'Monarch Biolink',
'predicate': ['blm:similar_to']
}
print("""Mod1B OwlSim Phenotype Similarity metadata:""")
pprint(self.meta)
class OwlsimSimilarity(object):
def __init__(self):
self.blw = BioLinkWrapper()
self.gene_set = []
self.input_object = ''
self.meta = {
'input_type': {
'complexity': 'set',
'id_type': 'HGNC',
'data_type': 'gene',
},
'output_type': {
'complexity': 'set',
'id_type': 'HGNC',
'data_type': 'gene',
},
'source': 'Monarch Biolink',
'predicate': ['blm:similar_to']
}
print("""Mod1B OwlSim Phenotype Similarity metadata:""")
pprint(self.meta)
def load_input_object(self, input_object):
self.input_object = input_object
def load_gene_set(self):
for gene in self.input_object['input']:
self.gene_set.append({
'input_id': gene['hit_id'],
'sim_input_curie': gene['hit_id'],
'input_symbol': gene['hit_symbol']
})
def calculate_similarity(self):
results = []
for index, gene in enumerate(self.gene_set):
try:
owlsim = self.blw.compute_owlsim(gene['sim_input_curie'])
for match in owlsim['matches']:
if match['type'] == 'gene':
results.append({
'input_id': gene['input_id'],
'input_symbol': gene['input_symbol'],
'hit_symbol': match['label'],
'hit_id': match['id'],
'score': match['score'],
})
except Exception as e:
print(match['id'], e)
return results
class GeneInteractions(object):
def __init__(self):
self.blw = BioLinkWrapper(Config().get_biolink_api_endpoint())
self.meta = {
'input_type': {
'complexity': 'set',
'id_type': 'HGNC',
'data_type': 'gene',
},
'output_type': {
'complexity': 'set',
'id_type': 'HGNC',
'data_type': 'gene',
},
'source': 'Monarch Biolink',
'predicate': ['blm:interacts with']
}
def metadata(self):
print("""Mod1E Interaction Network metadata:""")
pprint(self.meta)
@staticmethod
# RMB: July 5, 2019 - gene_records is a Pandas DataFrame
def load_gene_set(gene_records):
annotated_gene_set = []
for gene in gene_records.to_dict(orient='records'):
annotated_gene_set.append({
'input_id': gene['hit_id'],
'sim_input_curie': gene['hit_id'],
'input_symbol': gene['hit_symbol']
})
return annotated_gene_set
def get_interactions(self, input_gene_set, threshold):
annotated_input_gene_set = self.load_gene_set(input_gene_set)
lower_bound = int(threshold)
results = []
for gene in annotated_input_gene_set:
interactions = self.blw.gene_interactions(gene_curie=gene['sim_input_curie'])
for assoc in interactions['associations']:
interaction = \
self.blw.parse_association(
input_id=gene['sim_input_curie'],
input_label=gene['input_symbol'],
association=assoc
)
results.append({
'input_id': interaction['input_id'],
'input_symbol': interaction['input_symbol'],
'hit_symbol': interaction['hit_symbol'],
'hit_id': interaction['hit_id'],
'score': 0,
})
# Process the results
results = pd.DataFrame().from_records(results)
counts = results['hit_symbol'].value_counts().rename_axis('unique_values').to_frame('counts').reset_index()
high_counts = counts[counts['counts'] > lower_bound]['unique_values'].tolist()
results = pd.DataFrame(results[results['hit_symbol'].isin(high_counts)])
return results