Darwin passing build 1.

chp_client/query.py

@@ -144,10 +144,10 @@ def build_wildcard_query(
     return query
 
 def build_onehop_query(
-        q_subject,
-        q_subject_category,
-        q_object,
-        q_object_category,
+        q_subjects,
+        q_subject_categories,
+        q_objects,
+        q_object_categories,
         genes=None,
         drugs=None,
         outcome=None,
@@ -163,21 +163,21 @@ def build_onehop_query(
     q = message.query_graph
 
     # Add nodes
-    subject_node = q.add_node(q_subject, q_subject_category)
-    object_node = q.add_node(q_object, q_object_category)
+    subject_node = q.add_node(q_subjects, q_subject_categories)
+    object_node = q.add_node(q_objects, q_object_categories)
 
     # Add edge
     try:
-        edge_predicate = SUBJECT_TO_OBJECT_PREDICATE_MAP[(q_subject_category, q_object_category)]
+        edge_predicate = SUBJECT_TO_OBJECT_PREDICATE_MAP[(q_subject_categories[0], q_object_categories[0])]
     except KeyError:
-        raise QueryBuildError('Edge from {} to {} is not supported.'.format(q_subject_category, q_object_category))
+        raise QueryBuildError('Edge from {} to {} is not supported.'.format(q_subject_categories[0], q_object_categories[0]))
 
     edge_id = q.add_edge(subject_node, object_node, edge_predicate)
 
     # Add constraints
     if outcome is not None:
-        q.add_constraint('predicate_proxy', 'CHP:PredicateProxy', '==', outcome_name, edge_id=edge_id)
-        q.add_constraint(outcome_name, outcome, outcome_op, outcome_value, edge_id=edge_id)
+        q.add_constraint('predicate_proxy', 'CHP:PredicateProxy', '==', [outcome], edge_id=edge_id)
+        q.add_constraint(outcome, outcome, outcome_op, outcome_value, edge_id=edge_id)
 
     # Get context
     context = []

utils/build_batch_one_hop_queries.py

@@ -0,0 +1,132 @@
+
+from chp_client.query import build_onehop_query
+import itertools
+import tqdm
+import logging
+import pickle
+import random
+import json
+from collections import defaultdict
+
+from trapi_model.constants import *
+from chp.trapi_interface import TrapiInterface
+
+logging.basicConfig(level=logging.INFO)
+logger = logging.getLogger(__name__)
+logger.setLevel(logging.INFO)
+
+# Set number of queries to build
+NUM_QUERIES = 10
+
+TRAPI_VERSIONS = ['1.0', '1.1']
+
+# Set seed
+random.seed(111)
+
+# Get client
+#client = get_client()
+
+# Get curies
+logger.info('Getting curies.')
+curies = TrapiInterface().get_curies()
+#curies = client.curies()
+logger.info('Got curies.')
+
+# Build all simple single gene, single drug, breast cancer, survival queries.
+queries = defaultdict(lambda: defaultdict(list))
+disease = ['MONDO:0007254']
+outcome_name = 'survival_time'
+outcome = 'EFO:0000714'
+outcome_op = '>'
+
+# Build batch gene to disease query
+for trapi_version in TRAPI_VERSIONS:
+    genes = [gene for gene in random.choices(list(curies['biolink:Gene'].keys()), k=random.randint(1,4))]
+    q = build_onehop_query(
+            genes,
+            [BIOLINK_GENE],
+            disease,
+            [BIOLINK_DISEASE],
+            outcome=outcome,
+            outcome_op=outcome_op,
+            outcome_value=random.randint(500,1500),
+            trapi_version=trapi_version,
+            )
+    print(q)
+    input()
+    queries[trapi_version]['gene_to_disease_proxy'] = q.to_dict()
+
+
+# Build batch drug to disease query
+    drugs = [gene for gene in random.choices(list(curies['biolink:Drug'].keys()), k=random.randint(1,4))]
+    q = build_onehop_query(
+            drugs,
+            [BIOLINK_DRUG],
+            disease,
+            [BIOLINK_DISEASE],
+            outcome=outcome,
+            outcome_op=outcome_op,
+            outcome_value=random.randint(500,1500),
+            trapi_version=trapi_version,
+            )
+    print(q)
+    input()
+    queries[trapi_version]['drug_to_disease_proxy'] = q.to_dict()
+
+# Build batch gene to drug query
+    genes = [gene for gene in random.choices(list(curies['biolink:Gene'].keys()), k=random.randint(1,4))]
+    drugs = [gene for gene in random.choices(list(curies['biolink:Drug'].keys()), k=random.randint(1,4))]
+    q = build_onehop_query(
+            genes,
+            [BIOLINK_GENE],
+            drugs,
+            [BIOLINK_DRUG],
+            outcome=outcome,
+            outcome_op=outcome_op,
+            outcome_value=random.randint(500,1500),
+            trapi_version=trapi_version,
+            )
+    print(q)
+    input()
+    queries[trapi_version]['gene_to_drug_proxy'] = q.to_dict()
+
+
+# Build gene to drug query
+    genes = [gene for gene in random.choices(list(curies['biolink:Gene'].keys()), k=random.randint(1,4))]
+    drugs = [gene for gene in random.choices(list(curies['biolink:Drug'].keys()), k=random.randint(1,4))]
+    q = build_onehop_query(
+            drugs,
+            [BIOLINK_DRUG],
+            genes,
+            [BIOLINK_GENE],
+            outcome=outcome,
+            outcome_op=outcome_op,
+            outcome_value=random.randint(500,1500),
+            trapi_version=trapi_version,
+            )
+    print(q)
+    input()
+    queries[trapi_version]['drug_to_gene_proxy'] = q.to_dict()
+
+# Build gene to disease with proxy and context
+    genes = [gene for gene in random.choices(list(curies['biolink:Gene'].keys()), k=random.randint(1,4))]
+    drugs = [gene for gene in random.choices(list(curies['biolink:Drug'].keys()), k=random.randint(1,4))]
+    q = build_onehop_query(
+            genes,
+            [BIOLINK_GENE],
+            disease,
+            [BIOLINK_DISEASE],
+            outcome=outcome,
+            outcome_op=outcome_op,
+            outcome_value=random.randint(500,1500),
+            trapi_version=trapi_version,
+            drugs=drugs,
+            )
+    print(q)
+    input()
+    queries[trapi_version]['gene_to_disease_proxy_context'] = q.to_dict()
+
+
+# Pickle the queries
+with open('standard_batch_onehop_queries.pk', 'wb') as f_:
+    pickle.dump(dict(queries), f_)

utils/build_batch_wildcard_onehop_queries.py

@@ -0,0 +1,64 @@
+from chp_client.query import build_onehop_query
+import itertools
+import tqdm
+import logging
+import pickle
+import random
+import json
+from collections import defaultdict
+
+from trapi_model.constants import *
+from chp.trapi_interface import TrapiInterface
+
+logging.basicConfig(level=logging.INFO)
+logger = logging.getLogger(__name__)
+logger.setLevel(logging.INFO)
+
+# Set number of queries to build
+NUM_QUERIES = 10
+
+TRAPI_VERSIONS = ['1.0', '1.1']
+
+# Set seed
+random.seed(111)
+
+# Get client
+#client = get_client()
+
+# Get curies
+logger.info('Getting curies.')
+curies = TrapiInterface().get_curies()
+#curies = client.curies()
+logger.info('Got curies.')
+
+# Build all simple single gene, single drug, breast cancer, survival queries.
+queries = defaultdict(lambda: defaultdict(list))
+disease = ['MONDO:0007254']
+outcome_name = 'survival_time'
+outcome = 'EFO:0000714'
+outcome_op = '>'
+
+# Build batch gene to disease query
+for trapi_version in TRAPI_VERSIONS:
+# Build batch gene wildcard and drug wildcard to disease query
+    drug = [random.choice(list(curies["biolink:Drug"].keys()))]
+    gene = [random.choice(list(curies["biolink:Gene"].keys()))]
+    q = build_onehop_query(
+            None,
+            [BIOLINK_GENE, BIOLINK_DRUG],
+            disease,
+            [BIOLINK_DISEASE],
+            outcome=outcome,
+            outcome_op=outcome_op,
+            outcome_value=random.randint(500,1500),
+            trapi_version=trapi_version,
+            drugs=drug,
+            genes=gene,
+            )
+    print(q)
+    input()
+    queries[trapi_version]['batchwildcard_to_disease_proxy'] = q.to_dict()
+
+# Pickle the queries
+with open('wildcard_batch_onehop_queries.pk', 'wb') as f_:
+    pickle.dump(dict(queries), f_)

utils/build_drug_wildcard_batch_queries.py

@@ -5,6 +5,7 @@
 import pickle
 import random
 import json
+from collections import defaultdict
 
 from chp.trapi_interface import TrapiInterface
 
@@ -15,6 +16,8 @@
 # Set number of queries to build
 NUM_QUERIES = 10
 
+TRAPI_VERSIONS = ['1.0', '1.1']
+
 # Set seed
 random.seed(111)
 
@@ -26,24 +29,25 @@
 curies = TrapiInterface().get_curies()
 #curies = client.curies()
 logger.info('Got curies.')
-# Build all simple single gene, single drug, breast cancer, survival queries.
 
-queries = []
-for _ in range(NUM_QUERIES):
-    genes = [gene for gene in random.choices(list(curies['biolink:Gene'].keys()), k=random.randint(1,4))]
-    q = build_wildcard_query(
-        batch_genes=genes,
-        disease='MONDO:0007254',
-        outcome_name='survival_time',
-        outcome='EFO:0000714', 
-        outcome_op='>=',
-        outcome_value=random.randint(1, 5000),
-        trapi_version='1.0',
-        wildcard_category='drug',
-        )
-    #print(q)
-    #input()
-    queries.append(q.to_dict())
+# Build all simple single gene, single drug, breast cancer, survival queries.
+queries = defaultdict(list)
+for trapi_version in TRAPI_VERSIONS:
+    for _ in range(NUM_QUERIES):
+        genes = [gene for gene in random.choices(list(curies['biolink:Gene'].keys()), k=random.randint(1,4))]
+        q = build_wildcard_query(
+            batch_genes=genes,
+            disease='MONDO:0007254',
+            outcome_name='survival_time',
+            outcome='EFO:0000714', 
+            outcome_op='>',
+            outcome_value=random.randint(1, 5000),
+            trapi_version=trapi_version,
+            wildcard_category='drug',
+            )
+        #print(q)
+        #input()
+        queries[trapi_version].append(q.to_dict())
 
 # Pickle the queries
 with open('random_drug_wildcard_batch_queries.pk', 'wb') as f_:

utils/build_drug_wildcard_queries.py

@@ -5,6 +5,7 @@
 import pickle
 import random
 import json
+from collections import defaultdict
 
 from chp.trapi_interface import TrapiInterface
 
@@ -15,6 +16,8 @@
 # Set number of queries to build
 NUM_QUERIES = 10
 
+TRAPI_VERSIONS = ['1.0', '1.1']
+
 # Set seed
 random.seed(111)
 
@@ -26,22 +29,23 @@
 curies = TrapiInterface().get_curies()
 #curies = client.curies()
 logger.info('Got curies.')
-# Build all simple single gene, single drug, breast cancer, survival queries.
 
-queries = []
-for _ in range(NUM_QUERIES):
-    genes = [gene for gene in random.choices(list(curies['biolink:Gene'].keys()), k=random.randint(1,2))]
-    q = build_wildcard_query(
-        genes=genes,
-        disease='MONDO:0007254',
-        outcome_name='survival_time',
-        outcome='EFO:0000714', 
-        outcome_op='>=',
-        outcome_value=random.randint(1, 5000),
-        trapi_version='1.0',
-        wildcard_category='drug',
-        )
-    queries.append(q.to_dict())
+# Build all simple single gene, single drug, breast cancer, survival queries.
+queries = defaultdict(list)
+for trapi_version in TRAPI_VERSIONS:
+    for _ in range(NUM_QUERIES):
+        genes = [gene for gene in random.choices(list(curies['biolink:Gene'].keys()), k=random.randint(1,2))]
+        q = build_wildcard_query(
+            genes=genes,
+            disease='MONDO:0007254',
+            outcome_name='survival_time',
+            outcome='EFO:0000714', 
+            outcome_op='>',
+            outcome_value=random.randint(1, 5000),
+            trapi_version=trapi_version,
+            wildcard_category='drug',
+            )
+        queries[trapi_version].append(q.to_dict())
 
 # Pickle the queries
 with open('random_drug_wildcard_queries.pk', 'wb') as f_: