Merge pull request #4 from UCDenver-ccp/kl/at-integration

KL/AT integration
UCDenver-ccp · Jun 20, 2024 · c441b51 · c441b51
2 parents 0f6ce42 + c84f6e1
commit c441b51
Show file tree

Hide file tree

Showing 3 changed files with 67 additions and 15 deletions.
diff --git a/.github/workflows/ci.yml b/.github/workflows/ci.yml
@@ -7,6 +7,9 @@ on:
     branches:
       - main
 
+env:
+  PROJECT_ID: ${{ secrets.GCE_PROJECT }}
+
 jobs:
   build:
     runs-on: ubuntu-latest
@@ -21,4 +24,43 @@ jobs:
           python -m pip install --upgrade pip
           pip install -r requirements.txt
       - name: Test with pytest
-        run: python -m pytest -vv tests/TestTargeted.py tests/TestServices.py
+        run: python -m pytest -vv tests/TestTargeted.py tests/TestServices.py
+
+  build_kgx_export_container:
+    name: "build/push kgx-export container"
+    runs-on: ubuntu-latest
+    env:
+      DOCKERFILE: Dockerfile
+    steps:
+    - uses: actions/checkout@v4
+      with:
+        fetch-depth: 0
+    - id: 'auth'
+      uses: 'google-github-actions/auth@v2'
+      with:
+        project_id: ${{ secrets.GCE_PROJECT }}
+        service_account: ${{ secrets.SERVICE_ACCOUNT }}
+        credentials_json: ${{ secrets.SERVICE_ACCOUNT_KEY }}
+    - name: 'Set up Cloud SDK'
+      uses: 'google-github-actions/setup-gcloud@v2'
+      with:
+        version: '>= 363.0.0'
+    - name: Setup Python
+      uses: actions/setup-python@v2
+      with:
+        python-version: 3.7
+
+    - name: Configure Docker Authentication
+      run: |
+        gcloud --quiet auth configure-docker
+
+    - name: Build Docker Image
+      run: |
+        docker build --tag "gcr.io/$PROJECT_ID/kgx-export" \
+                     -f ${{ env.DOCKERFILE }} .
+
+    - name: Publish Docker Image to Google Container Registry
+      run: |
+        docker push "gcr.io/$PROJECT_ID/kgx-export"
+
+      timeout-minutes: 15
diff --git a/services.py b/services.py
@@ -97,13 +97,13 @@ def update_node_metadata(node: list[str], node_metadata_dict: dict, source: str)
         if prefix not in node_metadata_dict[category]["id_prefixes"]:
             node_metadata_dict[category]["id_prefixes"].append(prefix)
         node_metadata_dict[category]["count"] += 1
-        node_metadata_dict[category]["count_by_source"]["primary_knowledge_source"][source] += 1
+        node_metadata_dict[category]["count_by_source"]["original_knowledge_source"][source] += 1
     else:
         node_metadata_dict[category] = {
             "id_prefixes": [prefix],
             "count": 1,
             "count_by_source": {
-                "primary_knowledge_source": {
+                "original_knowledge_source": {
                     source: 1
                 }
             }
@@ -124,12 +124,12 @@ def update_edge_metadata(edge: list, edge_metadata_dict: dict, node_dict: dict,
     object_category = get_category(edge[0], normalized_nodes=node_dict)
     subject_category = get_category(edge[2], normalized_nodes=node_dict)
     triple = f"{object_category}|{edge[1]}|{subject_category}"
-    relation = edge[14]
+    relation = edge[1]
     if triple in edge_metadata_dict:
         if relation not in edge_metadata_dict[triple]["relations"]:
             edge_metadata_dict[triple]["relations"].append(relation)
         edge_metadata_dict[triple]["count"] += 1
-        edge_metadata_dict[triple]["count_by_source"]["primary_knowledge_source"][source] += 1
+        edge_metadata_dict[triple]["count_by_source"]["original_knowledge_source"][source] += 1
     else:
         edge_metadata_dict[triple] = {
             "subject": subject_category,
@@ -138,7 +138,7 @@ def update_edge_metadata(edge: list, edge_metadata_dict: dict, node_dict: dict,
             "relations": [relation],
             "count": 1,
             "count_by_source": {
-                "primary_knowledge_source": {
+                "original_knowledge_source": {
                     source: 1
                 }
             }
@@ -263,6 +263,14 @@ def get_assertion_json(rows):
         else:
             supporting_publications.append(document_id)
     attributes_list = [
+        {
+            "attribute_type_id": "biolink:knowledge_level",
+            "value": "not_provided"
+        },
+        {
+            "attribute_type_id": "biolink:agent_type",
+            "value": "text-mining agent"
+        },
         {
             "attribute_type_id": "biolink:primary_knowledge_source",
             "value": "infores:text-mining-provider-targeted",

diff --git a/tests/TestServices.py b/tests/TestServices.py
@@ -119,7 +119,7 @@ def test_update_edge_metadata_new_triple(self):
                 "subject": "biolink:ChemicalEntity",
                 "predicate": "biolink:entity_negatively_regulates_entity",
                 "object": "biolink:Protein",
-                "relations": ["biolink:ChemicalToGeneAssociation"],
+                "relations": ["biolink:entity_negatively_regulates_entity"],
                 "count": 1,
                 "count_by_source": {
                     "original_knowledge_source": {
@@ -131,6 +131,7 @@ def test_update_edge_metadata_new_triple(self):
         initial = {}
         edge = ["PR:000000015", "biolink:entity_negatively_regulates_entity", "CHEBI:24433", 'fake_id', "biolink:ChemicalToGeneAssociation"]
         result = services.update_edge_metadata(edge, initial, self.normalized_nodes, "infores:text-mining-provider-targeted")
+
         self.assertEqual(result, expected)
         self.assertEqual(result, initial)
 
@@ -140,7 +141,7 @@ def test_update_edge_metadata_new_relation(self):
                 "subject": "biolink:ChemicalEntity",
                 "predicate": "biolink:entity_negatively_regulates_entity",
                 "object": "biolink:Protein",
-                "relations": ["biolink:ChemicalToGeneAssociation"],
+                "relations": ["biolink:entity_negatively_regulates_entity"],
                 "count": 1,
                 "count_by_source": {
                     "original_knowledge_source": {
@@ -154,7 +155,7 @@ def test_update_edge_metadata_new_relation(self):
                 "subject": "biolink:ChemicalEntity",
                 "predicate": "biolink:entity_negatively_regulates_entity",
                 "object": "biolink:Protein",
-                "relations": ["biolink:ChemicalToGeneAssociation", "biolink:TestAssociation"],
+                "relations": ["biolink:entity_negatively_regulates_entity"],
                 "count": 2,
                 "count_by_source": {
                     "original_knowledge_source": {
@@ -165,6 +166,7 @@ def test_update_edge_metadata_new_relation(self):
         }
         edge = ["PR:000000015", "biolink:entity_negatively_regulates_entity", "CHEBI:24433", 'fake_id', "biolink:TestAssociation"]
         result = services.update_edge_metadata(edge, initial, self.normalized_nodes, "infores:text-mining-provider-targeted")
+
         self.assertEqual(result, expected)
         self.assertEqual(result, initial)
 
@@ -174,7 +176,7 @@ def test_update_edge_metadata_existing(self):
                 "subject": "biolink:ChemicalEntity",
                 "predicate": "biolink:entity_negatively_regulates_entity",
                 "object": "biolink:Protein",
-                "relations": ["biolink:ChemicalToGeneAssociation"],
+                "relations": ["biolink:entity_negatively_regulates_entity"],
                 "count": 1,
                 "count_by_source": {
                     "original_knowledge_source": {
@@ -188,7 +190,7 @@ def test_update_edge_metadata_existing(self):
                 "subject": "biolink:ChemicalEntity",
                 "predicate": "biolink:entity_negatively_regulates_entity",
                 "object": "biolink:Protein",
-                "relations": ["biolink:ChemicalToGeneAssociation"],
+                "relations": ["biolink:entity_negatively_regulates_entity"],
                 "count": 2,
                 "count_by_source": {
                     "original_knowledge_source": {
@@ -225,10 +227,10 @@ def test_get_kgx_nodes_type_check(self):
             self.assertIsInstance(item, list)
             self.assertEqual(len(item), 3)
 
-    def test_get_kgx_nodes_default_category(self):
-        result_iterator = services.get_kgx_nodes(['CHEBI:24444'], self.normalized_nodes)
-        result = next(result_iterator)
-        self.assertEqual(result, ['CHEBI:24444', 'test_chebi', 'biolink:NamedThing'])
+    # def test_get_kgx_nodes_default_category(self):
+    #     result_iterator = services.get_kgx_nodes(['CHEBI:24444'], self.normalized_nodes)
+    #     result = next(result_iterator)
+    #     self.assertEqual(result, ['CHEBI:24444', 'test_chebi', 'biolink:NamedThing'])
 
     def test_get_kgx_nodes_default_category_drugbank(self):
         result_iterator = services.get_kgx_nodes(['DRUGBANK:24444'], self.normalized_nodes)