def test_negative(self):
# empty response
reasoner_std = { "query_graph": dict(),
"knowledge_graph": dict(),
"response": dict()
}
# empty query graph
reasoner_std["query_graph"] = { "edges": [],
"nodes": []
}
# empty knowledge graph
reasoner_std["knowledge_graph"] = { "edges": [],
"nodes": []
}
# empty response graph
reasoner_std["results"] = { "node_bindings": [],
"edge_bindings": []
}
# add in evidence drug
drug = ('CYCLOPHOSPHAMIDE', 'CHEMBL:CHEMBL88')
reasoner_std['query_graph']['nodes'].append({ 'id':'n{}'.format('0'),
'type':'Drug',
'curie':'{}'.format(drug[1])
})
# add target survival node
phenotype = ('Survival_Time', 'EFO:0000714')
reasoner_std['query_graph']['nodes'].append({ 'id': 'n{}'.format('1'),
'type': 'PhenotypicFeature',
'curie': '{}'.format(phenotype[1]),
})
# link evidence to target survival node
reasoner_std['query_graph']['edges'].append({ 'id':'e{}'.format('0'),
'type':'chemical_to_disease_or_phenotypic_feature_association',
'value':1000,
'source_id':'n{}'.format('0'),
'target_id':'n{}'.format('1')
})
handler = ReasonerStdHandler(source_ara='relay9_22',
dict_query=reasoner_std)
queries = handler.buildChpQueries()
queries = handler.runChpQueries()
reasoner_std_final = handler.constructDecoratedKG()
KG = reasoner_std_final['knowledge_graph']
res = reasoner_std_final['results']
# extract probability
KG_result_edge = res['edge_bindings'][0]['kg_id']
for edge in KG['edges']:
if edge['type'] == 'chemical_to_disease_or_phenotypic_feature_association':
p_survival = edge['has_confidence_level']
gene_contribs = edge['Description']
print("probability of survival:",p_survival)
print(gene_contribs)
def test_contribution_property(self):
""" check if the contribution property returns contributions
"""
# empty response
reasoner_std = { "query_graph": dict(),
"knowledge_graph": dict(),
"results": list()
}
# empty query graph
reasoner_std["query_graph"] = { "edges": dict(),
"nodes": dict()
}
# empty knowledge graph
reasoner_std["knowledge_graph"] = { "edges": dict(),
"nodes": dict()
}
# add in evidence gene
gene1 = ('RAF1', 'ENSEMBL:ENSG00000132155')
reasoner_std['query_graph']['nodes']['n0'] = { 'category':'biolink:Gene',
'id':'{}'.format(gene1[1])
}
# add in disease node
disease = ('Breast_Cancer', 'MONDO:0007254')
reasoner_std['query_graph']['nodes']['n1'] = { 'category':'biolink:Disease',
'id':'{}'.format(disease[1])
}
# add target survival node
phenotype = ('Survival_Time', 'EFO:0000714')
reasoner_std['query_graph']['nodes']['n2'] = { 'category': 'biolink:PhenotypicFeature',
'id': '{}'.format(phenotype[1]),
}
# link genes/drugs to disease
reasoner_std['query_graph']['edges']['e0'] = { 'predicate':'biolink:GeneToDiseaseAssociation',
'subject': 'n0',
'object': 'n1'
}
# link disease to target
reasoner_std['query_graph']['edges']['e1'] = { 'predicate':'biolink:DiseaseToPhenotypicFeatureAssociation',
'subject': 'n1',
'object': 'n2',
'properties': { 'qualifier':'>=',
'days':970,
'contributions':True
}
}
handler = ReasonerStdHandler(source_ara='default',
dict_query=reasoner_std)
queries = handler.buildChpQueries()
queries = handler.runChpQueries()
reasoner_std_final = handler.constructDecoratedKG()
KG = reasoner_std_final['message']['knowledge_graph']
for edge_key in KG['edges'].keys():
edge = KG['edges'][edge_key]
if edge['predicate'] == 'biolink:DiseaseToPhenotypicFeatureAssociation':
p_survival = edge['has_confidence_level']
report = edge['properties']['contributions']
print("probability of survival:",p_survival)
def test_gene_request(self):
# empty response
reasoner_std = {
"query_graph": dict(),
"knowledge_graph": dict(),
"response": dict()
}
# empty query graph
reasoner_std["query_graph"] = {"edges": [], "nodes": []}
# empty knowledge graph
reasoner_std["knowledge_graph"] = {"edges": [], "nodes": []}
# empty response graph
reasoner_std["results"] = {"node_bindings": [], "edge_bindings": []}
# add in evidence genes
gene = ('RAF1', 'ENSEMBL:ENSG00000132155')
reasoner_std['query_graph']['nodes'].append({
'id':
'n{}'.format('0'),
'type':
'Gene',
'curie':
'{}'.format(gene[1])
})
# add target survival node
reasoner_std['query_graph']['nodes'].append({
'id': 'n{}'.format('1'),
'type': 'PhenotypicFeature',
'curie': 'UBERON:0000071',
})
# link evidence to target survival node
reasoner_std['query_graph']['edges'].append({
'id':
'e{}'.format('0'),
'type':
'causes',
'onset_qualifier':
1000,
'source_id':
'n{}'.format('0'),
'target_id':
'n{}'.format('1')
})
handler = ReasonerStdHandler(source_ara='ranking',
dict_query=reasoner_std)
queries = handler.buildChpQueries()
queries = handler.runChpQueries()
reasoner_std_final = handler.constructDecoratedKG()
KG = reasoner_std_final['knowledge_graph']
res = reasoner_std_final['results']
# extract probability
KG_result_edge = res['edge_bindings'][0]['kg_id']
for edge in KG['edges']:
if edge['id'] == KG_result_edge:
p_survival = edge['has_confidence_level']
print("probability of survival:", p_survival)
def test_one_drug(self):
""" Test with one drug
"""
# empty response
reasoner_std = { "query_graph": dict(),
"knowledge_graph": dict(),
"results": list()
}
# empty query graph
reasoner_std["query_graph"] = { "edges": dict(),
"nodes": dict()
}
# empty knowledge graph
reasoner_std["knowledge_graph"] = { "edges": dict(),
"nodes": dict()
}
# add in evidence drug
drug = ('CYCLOPHOSPHAMIDE', 'CHEMBL:CHEMBL88')
reasoner_std['query_graph']['nodes']['n0'] = { 'category':'biolink:Drug',
'id':'{}'.format(drug[1])
}
# add in disease node
disease = ('Breast_Cancer', 'MONDO:0007254')
reasoner_std['query_graph']['nodes']['n1'] = { 'category':'biolink:Disease',
'id':'{}'.format(disease[1])
}
# add target survival node
phenotype = ('Survival_Time', 'EFO:0000714')
reasoner_std['query_graph']['nodes']['n2'] = { 'category': 'biolink:PhenotypicFeature',
'id': '{}'.format(phenotype[1]),
}
# link genes/drugs to disease
reasoner_std['query_graph']['edges']['e0'] = { 'predicate':'biolink:ChemicalToDiseaseOrPhenotypicFeatureAssociation',
'subject': 'n0',
'object': 'n1'
}
# link disease to target
reasoner_std['query_graph']['edges']['e1'] = { 'predicate':'biolink:DiseaseToPhenotypicFeatureAssociation',
'subject': 'n1',
'object': 'n2',
'properties': { 'qualifier':'>=',
'days':970
}
}
handler = ReasonerStdHandler(source_ara='default',
dict_query=reasoner_std)
queries = handler.buildChpQueries()
queries = handler.runChpQueries()
reasoner_std_final = handler.constructDecoratedKG()
KG = reasoner_std_final['message']['knowledge_graph']
for edge_key in KG['edges'].keys():
edge = KG['edges'][edge_key]
if edge['predicate'] == 'biolink:DiseaseToPhenotypicFeatureAssociation':
p_survival = edge['has_confidence_level']
print("probability of survival:",p_survival)
def test_no_evidence(self):
# empty response
reasoner_std = {
"query_graph": dict(),
"knowledge_graph": dict(),
"response": dict()
}
# empty query graph
reasoner_std["query_graph"] = {"edges": [], "nodes": []}
# empty knowledge graph
reasoner_std["knowledge_graph"] = {"edges": [], "nodes": []}
# empty response graph
reasoner_std["results"] = {"node_bindings": [], "edge_bindings": []}
# add in disease node
disease = ('Breast_Cancer', 'MONDO:0007254')
reasoner_std['query_graph']['nodes'].append({
'id':
'n{}'.format('0'),
'type':
'disease',
'curie':
'{}'.format(disease[1])
})
# add target survival node
phenotype = ('Survival_Time', 'EFO:0000714')
reasoner_std['query_graph']['nodes'].append({
'id':
'n{}'.format('1'),
'type':
'PhenotypicFeature',
'curie':
'{}'.format(phenotype[1]),
})
# link disease to target
reasoner_std['query_graph']['edges'].append({
'id':
'e{}'.format('0'),
'type':
'disease_to_phenotype_association',
'value':
970,
'source_id':
'n{}'.format('0'),
'target_id':
'n{}'.format('1')
})
handler = ReasonerStdHandler(source_ara='exploring',
dict_query=reasoner_std)
queries = handler.buildChpQueries()
queries = handler.runChpQueries()
reasoner_std_final = handler.constructDecoratedKG()
KG = reasoner_std_final['knowledge_graph']
for edge in KG['edges']:
if edge['type'] == 'disease_to_phenotype_association':
p_survival = edge['has_confidence_level']
print("probability of survival:", p_survival)
def test_no_evidence_omitting_KG_and_results(self):
with self.assertRaises(SystemExit) as se:
# empty response
reasoner_std = {
"query_graph": dict(),
}
# empty query graph
reasoner_std["query_graph"] = {"edges": [], "nodes": []}
# add in disease node
disease = ('Breast_Cancer', 'MONDO:0007254')
reasoner_std['query_graph']['nodes'].append({
'id':
'n{}'.format('0'),
'type':
'disease',
'curie':
'{}'.format(disease[1])
})
# add target survival node
phenotype = ('Survival_Time', 'EFO:0000714')
reasoner_std['query_graph']['nodes'].append({
'id':
'n{}'.format('1'),
'type':
'PhenotypicFeature',
'curie':
'{}'.format(phenotype[1]),
})
# link disease to target
reasoner_std['query_graph']['edges'].append({
'id':
'e{}'.format('0'),
'type':
'disease_to_phenotype_association',
'value':
970,
'source_id':
'n{}'.format('0'),
'target_id':
'n{}'.format('1')
})
handler = ReasonerStdHandler(source_ara='unsecret',
dict_query=reasoner_std)
queries = handler.buildChpQueries()
queries = handler.runChpQueries()
reasoner_std_final = handler.constructDecoratedKG()
KG = reasoner_std_final['knowledge_graph']
for edge in KG['edges']:
if edge['type'] == 'disease_to_phenotype_association':
p_survival = edge['has_confidence_level']
print("probability of survival:", p_survival)
self.assertEqual(se.exception.code, 'Needs at least 1 gene')
def test_sparse_drug(self):
""" queries the drug Gemzar which has a sparsity issue. Used to throw an error.
Should be handled.
"""
# empty response
reasoner_std = {
"query_graph": {},
"knowledge_graph": {},
"results": []
}
# empty query graph
reasoner_std["query_graph"] = {"edges": {}, "nodes": {}}
# empty knowledge graph
reasoner_std["knowledge_graph"] = {"edges": {}, "nodes": {}}
# add in evidence drug
drug = ('GEMZAR', 'CHEMBL:CHEMBL888')
reasoner_std['query_graph']['nodes']['n{}'.format('0')] = {
'category': 'biolink:Drug',
'id': '{}'.format(drug[1])
}
# add in gene node (to be filled by contribution analysis
reasoner_std['query_graph']['nodes']['n{}'.format('1')] = {
'category': 'biolink:Gene',
}
# link genes/drugs to disease
reasoner_std['query_graph']['edges']['e{}'.format(0)] = {
'predicate': 'biolink:chemical_to_gene_association',
'subject': 'n0',
'object': 'n1'
}
handler = ReasonerStdHandler(source_ara='default',
dict_query=reasoner_std,
max_results=5)
queries = handler.buildChpQueries()
queries = handler.runChpQueries()
reasoner_std_final = handler.constructDecoratedKG()
KG = reasoner_std_final["message"]['knowledge_graph']
res = reasoner_std_final["message"]['results']
res_pretty = json.dumps(reasoner_std_final, indent=2)
print(res_pretty)
def test_no_evidence_omitting_KG_and_results(self):
""" Test with no evidence, but omitting KG and Results (should be handled by handler)
"""
# empty response
reasoner_std = { "query_graph": dict(),
}
# empty query graph
reasoner_std["query_graph"] = { "edges": dict(),
"nodes": dict()
}
# add in disease node
disease = ('Breast_Cancer', 'MONDO:0007254')
reasoner_std['query_graph']['nodes']['n0'] = { 'category':'biolink:Disease',
'id':'{}'.format(disease[1])
}
# add target survival node
phenotype = ('Survival_Time', 'EFO:0000714')
reasoner_std['query_graph']['nodes']['n1'] = { 'category': 'biolink:PhenotypicFeature',
'id': '{}'.format(phenotype[1]),
}
# link disease to target
reasoner_std['query_graph']['edges']['e0'] = { 'predicate':'biolink:DiseaseToPhenotypicFeatureAssociation',
'subject': 'n0',
'object': 'n1',
'properties': { 'qualifier':'>=',
'days':970
}
}
handler = ReasonerStdHandler(source_ara='default',
dict_query=reasoner_std)
queries = handler.buildChpQueries()
queries = handler.runChpQueries()
reasoner_std_final = handler.constructDecoratedKG()
KG = reasoner_std_final['message']['knowledge_graph']
for edge_key in KG['edges'].keys():
edge = KG['edges'][edge_key]
if edge['predicate'] == 'biolink:DiseaseToPhenotypicFeatureAssociation':
p_survival = edge['has_confidence_level']
print("probability of survival:",p_survival)
def test_drug(self):
""" queries Cyclophosphamide which has no sparsity issue. Should return
top 5 genes and a probability of survival
"""
# empty response
reasoner_std = {
"query_graph": {},
"knowledge_graph": {},
"results": []
}
# empty query graph
reasoner_std["query_graph"] = {"edges": {}, "nodes": {}}
# empty knowledge graph
reasoner_std["knowledge_graph"] = {"edges": {}, "nodes": {}}
# add in evidence drug
drug = ('CYCLOPHOSPHAMIDE', 'CHEMBL:CHEMBL88')
reasoner_std['query_graph']['nodes']['n0'] = {
'category': 'biolink:Drug',
'id': '{}'.format(drug[1])
}
# add in gene node (to be filled by contribution analysis
reasoner_std['query_graph']['nodes']['n1'] = {
'category': 'biolink:Gene'
}
#add in disease node
disease = ('Breast_Cancer', 'MONDO:0007254')
reasoner_std['query_graph']['nodes']['n2'] = {
'category': 'biolink:Disease',
'id': '{}'.format(disease[1])
}
# add target survival node
phenotype = ('Survival_Time', 'EFO:0000714')
reasoner_std['query_graph']['nodes']['n3'] = {
'category': 'biolink:PhenotypicFeature',
'id': '{}'.format(phenotype[1]),
}
# link disease to target survival node
reasoner_std['query_graph']['edges']['e0'] = {
'predicate': 'biolink:DiseaseToPhenotypicFeatureAssociation',
'subject': 'n2',
'object': 'n3',
'properties': {
'qualifier': '>=',
'days': 940
}
}
# link genes/drugs to disease
reasoner_std['query_graph']['edges']['e1'] = {
'predicate': 'biolink:GeneToDiseaseAssociation',
'subject': 'n1',
'object': 'n2'
}
reasoner_std['query_graph']['edges']['e2'] = {
'predicate':
'biolink:ChemicalToDiseaseOrPhenotypicFeatureAssociation',
'subject': 'n0',
'object': 'n2'
}
handler = ReasonerStdHandler(source_ara='default',
dict_query=reasoner_std,
max_results=5)
queries = handler.buildChpQueries()
queries = handler.runChpQueries()
reasoner_std_final = handler.constructDecoratedKG()
KG = reasoner_std_final["message"]['knowledge_graph']
res_pretty = json.dumps(reasoner_std_final, indent=2)
print(res_pretty)
# extract probability
for _, edge in KG['edges'].items():
if edge['predicate'] == 'biolink:DiseaseToPhenotypicFeatureAssociation':
p_survival = edge['has_confidence_level']
#gene_contribs = edge['Description']
break
print("probability of survival:", p_survival)
def test_normal_two_genes_and_drug(self):
# empty response
reasoner_std = {
"query_graph": dict(),
"knowledge_graph": dict(),
"response": dict()
}
# empty query graph
reasoner_std["query_graph"] = {"edges": [], "nodes": []}
# empty knowledge graph
reasoner_std["knowledge_graph"] = {"edges": [], "nodes": []}
# empty response graph
reasoner_std["results"] = {"node_bindings": [], "edge_bindings": []}
# add in evidence gene
gene1 = ('RAF1', 'ENSEMBL:ENSG00000132155')
reasoner_std['query_graph']['nodes'].append({
'id':
'n{}'.format('0'),
'type':
'Gene',
'curie':
'{}'.format(gene1[1])
})
# add in second evidence gene
gene2 = ('BRCA1', 'ENSEMBL:ENSG00000012048')
reasoner_std['query_graph']['nodes'].append({
'id':
'n{}'.format('1'),
'type':
'Gene',
'curie':
'{}'.format(gene2[1])
})
# add in evidence drug
drug = ('CYCLOPHOSPHAMIDE', 'CHEMBL:CHEMBL88')
reasoner_std['query_graph']['nodes'].append({
'id':
'n{}'.format('2'),
'type':
'Drug',
'curie':
'{}'.format(drug[1])
})
# add in disease node
disease = ('Breast_Cancer', 'MONDO:0007254')
reasoner_std['query_graph']['nodes'].append({
'id':
'n{}'.format('3'),
'type':
'disease',
'curie':
'{}'.format(disease[1])
})
# add target survival node
phenotype = ('Survival_Time', 'EFO:0000714')
reasoner_std['query_graph']['nodes'].append({
'id':
'n{}'.format('4'),
'type':
'PhenotypicFeature',
'curie':
'{}'.format(phenotype[1]),
})
# link genes/drugs to disease
reasoner_std['query_graph']['edges'].append({
'id':
'e{}'.format('0'),
'type':
'gene_to_disease_association',
'source_id':
'n{}'.format('0'),
'target_id':
'n{}'.format('3')
})
reasoner_std['query_graph']['edges'].append({
'id':
'e{}'.format('1'),
'type':
'gene_to_disease_association',
'source_id':
'n{}'.format('1'),
'target_id':
'n{}'.format('3')
})
reasoner_std['query_graph']['edges'].append({
'id':
'e{}'.format('2'),
'type':
'chemical_to_disease_or_phenotypic_feature_association',
'source_id':
'n{}'.format('2'),
'target_id':
'n{}'.format('3')
})
# link disease to target
reasoner_std['query_graph']['edges'].append({
'id':
'e{}'.format('3'),
'type':
'disease_to_phenotype_association',
'value':
970,
'source_id':
'n{}'.format('3'),
'target_id':
'n{}'.format('4')
})
handler = ReasonerStdHandler(source_ara='exploring',
dict_query=reasoner_std)
queries = handler.buildChpQueries()
queries = handler.runChpQueries()
reasoner_std_final = handler.constructDecoratedKG()
KG = reasoner_std_final['knowledge_graph']
for edge in KG['edges']:
if edge['type'] == 'disease_to_phenotype_association':
p_survival = edge['has_confidence_level']
print("probability of survival:", p_survival)
'target_id':
'n{}'.format(nodeCount - 1)
})
edgeCount += 1
# BKB target
response['probability_targets'] = [('Survival_Time', '>=', 1000)]
with open('sample_query1.pk', 'wb') as f_:
pickle.dump(response, f_)
handler = ReasonerStdHandler(source_ara='unsecret', dict_query=response)
queries = handler.buildChpQueries()
queries = handler.runChpQueries()
reasoner_std_final = handler.constructDecoratedKG()
#print(reasoner_std_final)
'''
#-- Get all options to fill query graph edges
options = dict()
for edge in reasoner_std['query_graph']['edges']:
options[edge['id']] = []
for result in reasoner_std['results']['edge_bindings']:
for qg_id, kg_ids in result.items():
options[qg_id].append(kg_ids[0])
#-- Hash edges by ID
edges = {edge['id']: edge for edge in reasoner_std['knowledge_graph']['edges']}
nodes = {node['id']: node for node in reasoner_std['knowledge_graph']['nodes']}
#-- Get answers defined as a set of edges such that all nodes are consistent, i.e. complete subgraph that answers query.