def test_standard_api_resource(client, requests_mock):
assert ordered(StandardAPIResource.handle_exception(
[
Exception('foo'),
TranQLException('bar',details='baz')
],
warning=True
)) == ordered({
"errors" : [
{
"message" : "foo",
"details" : ""
},
{
"message" : "bar",
"details" : "baz"
}
],
"status" : "Warning"
})
assert StandardAPIResource.response({
'status' : 'Warning',
'errors' : [],
'knowledge_graph' : {
'nodes' : [],
'edges' : []
}
}) == (
{
'status' : 'Warning',
'errors' : [],
'knowledge_graph' : {
'nodes' : [],
'edges' : []
}
},
200
)
assert StandardAPIResource.response({
'status' : 'Error',
'errors' : []
}) == (
{
'status' : 'Error',
'errors' : []
},
500
)
def test_decorate_kg(client):
knowledge_graph = {
"edges": [{
"id": "e0",
"source_id": "n0",
"target_id": "n1",
"type": "targets"
}],
"nodes": [{
"id": "n0",
"type": "chemical_substance"
}, {
"id": "n1",
"type": "gene"
}]
}
expected = {
"edges": [{
"id": "e0",
"source_id": "n0",
"target_id": "n1",
"type": "targets",
"reasoner": ["robokop"],
"source_database": ["unknown"]
}],
"nodes": [{
"id": "n0",
"type": "chemical_substance",
"reasoner": ["robokop"]
}, {
"id": "n1",
"type": "gene",
"reasoner": ["robokop"]
}]
}
args = {'reasoners': ['robokop']}
response = client.post(f'/tranql/decorate_kg',
query_string=args,
data=json.dumps(knowledge_graph),
content_type='application/json')
assert ordered(response.json) == ordered(expected)
def test_rtx_convert_curies(client):
converted_to_rtx = RtxQuery.convert_curies_to_rtx({
"question_graph": {
"nodes": [{
"curie": "CHEMBL:CHEMBL3",
"id": "chemical_substance",
"type": "chemical_substance"
}],
"edges": []
}
})
assert ordered(converted_to_rtx) == ordered({
"question_graph": {
"nodes": [{
"curie": "CHEMBL.COMPOUND:CHEMBL3",
"id": "chemical_substance",
"type": "chemical_substance"
}],
"edges": []
}
})
converted_to_standard = RtxQuery.convert_curies_to_standard({
"knowledge_map": [{
"node_bindings": {
"chemical_substance": ["CHEMBL.COMPOUND:CHEMBL3"]
},
"edge_bindings": {
"foo": [""]
}
}],
"knowledge_graph": {
"nodes": [{
"id": "CHEMBL.COMPOUND:CHEMBL3"
}, {
"id": "FOOBAR"
}],
"edges": [{
"source_id": "CHEMBL.COMPOUND:CHEMBL3",
"target_id": "FOOBAR"
}]
},
"question_graph": {
"nodes": [{
"curie": "CHEMBL.COMPOUND:CHEMBL3",
"id": "chemical_substance",
"type": "chemical_substance"
}],
"edges": []
}
})
assert ordered(converted_to_standard) == ordered({
"knowledge_map": [{
"node_bindings": {
"chemical_substance": "CHEMBL:CHEMBL3"
},
"edge_bindings": {
"foo": [""]
}
}],
"knowledge_graph": {
"nodes": [{
"id": "CHEMBL:CHEMBL3"
}, {
"id": "FOOBAR"
}],
"edges": [{
"source_id": "CHEMBL:CHEMBL3",
"target_id": "FOOBAR"
}]
},
"question_graph": {
"nodes": [{
"curie": "CHEMBL:CHEMBL3",
"id": "chemical_substance",
"type": "chemical_substance"
}],
"edges": []
}
})
def test_ast_merge_results (requests_mock):
set_mock(requests_mock, "workflow-5")
""" Validate that
-- Results from the query plan are being merged together correctly
"""
print("test_ast_merge_answers ()")
tranql = TranQL ()
tranql.resolve_names = False
ast = tranql.parse ("""
SELECT cohort_diagnosis:disease->diagnoses:disease
FROM '/clinical/cohort/disease_to_chemical_exposure'
WHERE cohort_diagnosis = 'MONDO:0004979' --asthma
AND Sex = '0'
AND cohort = 'all_patients'
AND max_p_value = '0.5'
SET '$.knowledge_graph.nodes.[*].id' AS diagnoses
""")
select = ast.statements[0]
# What is the proper format for the name of a mock file? This should be made into one
mock_responses = [
{
'knowledge_graph': {
'nodes': [
{'id': 'CHEBI:28177', 'type': 'chemical_substance'},
{'id': 'HGNC:2597', 'type': 'gene'},
{
'id': 'egg',
'name':'test_name_merge',
'type': 'foo_type',
'test_attr': ['a','b']
},
{
'id': 'equivalent_identifier_merge',
'equivalent_identifiers': ['TEST:00000'],
'merged_property': [
'a',
'b'
]
}
],
'edges': [
{'id': 'e0', 'source_id': 'CHEBI:28177', 'target_id': 'HGNC:2597'},
{
# Test if edges that are connected to merged nodes will be successfully merged with other duplicate edges
'source_id' : 'CHEBI:28177',
'target_id' : 'egg',
'type': ['merge_this'],
'merge_this_list' : ['edge_1'],
'unique_attr_e_1' : 'e_1',
'id' : 'winning_edge_id'
},
]
},
'knowledge_map': [
{
'node_bindings': {
'chemical_substance': 'CHEBI:28177',
'gene': 'HGNC:2597'
},
'edge_bindings': {}
}
]
},
{
'knowledge_graph': {
'nodes': [
{'id': 'CHEBI:28177', 'type': 'chemical_substance'},
{
'id': 'also_test_array_type_and_string_type_merge',
'name':'test_name_merge',
'type': ['foo_type','bar_type'],
'test_attr': ['a','c']
},
{'id': 'TEST:00000', 'type': 'test', 'merged_property': ['a','c']},
],
'edges': [
{'id': 'e0', 'source_id': 'CHEBI:28177', 'target_id': 'TEST:00000'},
{
'source_id' : 'CHEBI:28177',
'target_id' : 'also_test_array_type_and_string_type_merge',
'type': ['merge_this'],
'merge_this_list' : ['edge_2'],
'unique_attr_e_2' : 'e_2'
}
]
},
'knowledge_map': [
{
'node_bindings': {
'chemical_substance': 'CHEBI:28177',
'test': 'TEST:00000'
},
'edge_bindings': {}
}
]
}
]
expected_result = {
"knowledge_graph": {
"edges": [
{
"id": "e0",
"source_id": "CHEBI:28177",
"target_id": "HGNC:2597",
"type": []
},
{
"id": "e0",
"source_id": "CHEBI:28177",
"target_id": "equivalent_identifier_merge",
"type": []
},
{
"id" : "winning_edge_id",
"source_id" : "CHEBI:28177",
"target_id" : "egg",
"type" : ["merge_this"],
"merge_this_list" : ["edge_1", "edge_2"],
"unique_attr_e_1" : "e_1",
"unique_attr_e_2" : "e_2"
}
],
"nodes": [
{
"equivalent_identifiers": [
"CHEBI:28177"
],
"id": "CHEBI:28177",
"type": ["chemical_substance"]
},
{
"equivalent_identifiers": [
"HGNC:2597"
],
"id": "HGNC:2597",
"type": ["gene"]
},
{
"equivalent_identifiers": [
"also_test_array_type_and_string_type_merge",
"egg"
],
"type": [
"foo_type",
"bar_type"
],
"id": "egg",
"name": "test_name_merge",
"test_attr": [
"a",
"b",
"c"
]
},
{
"equivalent_identifiers": [
"TEST:00000",
"equivalent_identifier_merge"
],
"merged_property": ["a", "b", "c"],
"id": "equivalent_identifier_merge",
"type": ["test"]
}
]
},
"knowledge_map": [
{
"edge_bindings": {},
"node_bindings": {
"chemical_substance": "CHEBI:28177",
"gene": "HGNC:2597"
}
},
{
"edge_bindings": {},
"node_bindings": {
"chemical_substance": "CHEBI:28177",
"test": "equivalent_identifier_merge"
}
}
],
'question_graph': {
'edges': [
{
'id': 'foo',
'type': 'test'
}
],
'nodes': [
{
'id': 'bar',
'type': 'bartest'
}
]
}
}
merged_results = select.merge_results (
mock_responses,
tranql,
{
'edges': [
{
'id': 'foo',
'type': 'test'
}
],
'nodes': [
{
'id': 'bar',
'type': 'bartest'
}
]
},
root_order=None
)
assert ordered(merged_results) == ordered(expected_result)
def test_ast_merge_knowledge_maps (requests_mock):
set_mock(requests_mock, "workflow-5")
tranql = TranQL ()
tranql.asynchronous = False
tranql.resolve_names = False
ast = tranql.parse ("""
select chemical_substance->disease->gene
from "/schema"
where chemical_substance="CHEMBL:CHEMBL3"
""")
# select = ast.statements[0]
# statements = select.plan (select.planner.plan (select.query))
# print(statements[0].query.order)
# (select.execute_plan(tranql))
responses = [
{
'knowledge_map' : [
{
'node_bindings' : {
'chemical_substance' : 'CHEBI:100',
'disease' : 'MONDO:50'
},
'edge_bindings' : {
'e0' : 'ROOT_EDGE'
}
}
],
'question_order' : ['chemical_substance','disease']
},
{
'knowledge_map' : [
{
'node_bindings' : {
'disease' : 'MONDO:50',
'gene' : 'HGNC:1',
'metabolite' : 'KEGG:C00017'
},
'edge_bindings' : {
'e1' : 'TEST_EDGE'
}
}
],
'question_order' : ['disease','gene','metabolite']
},
{
'knowledge_map' : [
{
'node_bindings' : {
'disease' : 'MONDO:50',
'gene' : 'HGNC:1',
'metabolite' : 'KEGG:FOOBAR'
},
'edge_bindings' : {
}
}
],
'question_order' : ['disease','gene','metabolite']
},
{
'knowledge_map' : [
{
'node_bindings' : {
'metabolite' : 'KEGG:FOOBAR',
'protein' : 'UniProtKB:TESTING'
},
'edge_bindings' : {
}
}
],
'question_order' : ['metabolite','protein']
},
{
'knowledge_map' : [
{
'node_bindings' : {
'metabolite' : 'KEGG:C00017',
'protein' : 'UniProtKB:Q9NZJ5'
},
'edge_bindings' : {
}
}
],
'question_order' : ['metabolite','protein']
}
]
merged = SelectStatement.connect_knowledge_maps(responses,[
'chemical_substance',
'disease',
'gene',
'metabolite',
'protein'
])
assert_lists_equal(ordered(merged), ordered([
{
"node_bindings" : {
"chemical_substance" : "CHEBI:100",
"disease" : "MONDO:50",
"gene" : "HGNC:1",
"metabolite" : "KEGG:FOOBAR",
"protein" : "UniProtKB:TESTING"
},
"edge_bindings" : {
"e0" : "ROOT_EDGE"
}
},
{
"node_bindings" : {
"chemical_substance" : "CHEBI:100",
"disease" : "MONDO:50",
"gene" : "HGNC:1",
"metabolite" : "KEGG:C00017",
"protein" : "UniProtKB:Q9NZJ5"
},
"edge_bindings" : {
"e0" : "ROOT_EDGE",
"e1" : "TEST_EDGE",
}
}
]))
def test_merge_messages(client):
messages = [{
"knowledge_graph": {
"edges": [{
"source_id": "TEST:CS1",
"target_id": "TEST:G1",
"type": "targets"
}],
"nodes": [{
"id": "TEST:CS1",
"type": "chemical_substance"
}, {
"id": "TEST:G1",
"type": "gene"
}]
},
"knowledge_map": []
}, {
"knowledge_graph": {
"edges": [{
"source_id": "TEST:CS2",
"target_id": "TEST:G2",
"type": "interacts_with"
}],
"nodes": [{
"equivalent_identifiers": ["TEST:CS1"],
"id": "TEST:merged",
"type": ["chemical_substance", "Drug"]
}, {
"id": "TEST:CS2",
"type": "chemical_substance"
}, {
"id": "TEST:G2",
"type": "gene"
}]
},
"knowledge_map": []
}]
expected = {
"knowledge_graph": {
"nodes": [{
"id": "TEST:CS1",
"type": ["chemical_substance", "Drug"],
"equivalent_identifiers": ["TEST:merged", "TEST:CS1"]
}, {
"id": "TEST:G1",
"type": ["gene"],
"equivalent_identifiers": ["TEST:G1"]
}, {
"id": "TEST:CS2",
"type": ["chemical_substance"],
"equivalent_identifiers": ["TEST:CS2"]
}, {
"id": "TEST:G2",
"type": ["gene"],
"equivalent_identifiers": ["TEST:G2"]
}],
"edges": [{
"source_id": "TEST:CS1",
"target_id": "TEST:G1",
"type": ["targets"]
}, {
"source_id": "TEST:CS2",
"target_id": "TEST:G2",
"type": ["interacts_with"]
}]
},
"knowledge_map": [],
"question_graph": {
"nodes": [],
"edges": []
}
}
args = {
'name_based_merging': True,
'resolve_names': False,
'question_graph': json.dumps({
"nodes": [],
"edges": []
})
}
response = client.post(f'/tranql/merge_messages',
query_string=args,
data=json.dumps(messages),
content_type='application/json')
assert ordered(response.json) == ordered(expected)