Merge pull request #603 from monarch-initiative/issue-830-add-go-publ…

…ications

Populate publications for GO annotation ingest, plus update tests & doc to match

docs/Sources/go.md

@@ -43,6 +43,7 @@ __**Associations**__
     * object (go_term.id)
     * negated
     * has_evidence
+    * publications
     * aggregating_knowledge_source (["infores:monarchinitiative"])
     * primary_knowledge_source
 
@@ -55,6 +56,7 @@ OR
     * object (go_term.id)
     * negated
     * has_evidence
+    * publications
     * aggregating_knowledge_source (["infores:monarchinitiative"])
     * primary_knowledge_source
 
@@ -65,6 +67,7 @@ OR
     * object (go_term.id)
     * negated
     * has_evidence
+    * publications
     * aggregating_knowledge_source (["infores:monarchinitiative"])
     * primary_knowledge_source
 
@@ -75,6 +78,7 @@ OR
     * object (go_term.id)
     * negated
     * has_evidence
+    * publications
     * aggregating_knowledge_source (["infores:monarchinitiative"])
     * primary_knowledge_source
 
@@ -87,6 +91,7 @@ __**Possible Additional Gene to Gene Ontology Term Association?**__
     * object (go_term.id)
     * negated
     * has_evidence
+    * publications
     * aggregating_knowledge_source (["infores:monarchinitiative"])
     * primary_knowledge_source
 

src/monarch_ingest/ingests/go/annotation.py

@@ -18,7 +18,6 @@
 )
 from loguru import logger
 
-
 koza_app = get_koza_app("go_annotation")
 
 # for row in koza_app.source: # doesn't play nice with tests
@@ -101,6 +100,8 @@
             # actual primary knowledge source of the GOA knowledge statement
             assigned_by = get_infores(row['Assigned_By'])
 
+            publications = [p.replace('MGI:MGI:','MGI:') for p in row['DB_Reference'].split("|")] if row['DB_Reference'] else []
+
             # Instantiate the appropriate Gene-to-GO Term instance
             association = gene_go_term_association_class(
                 id="uuid:" + str(uuid.uuid1()),
@@ -109,6 +110,7 @@
                 predicate=predicate,
                 negated=negated,
                 has_evidence=[eco_term],
+                publications=publications,
                 # subject_context_qualifier=ncbitaxa,  # Biolink Pydantic model support missing for this slot
                 aggregator_knowledge_source=["infores:monarchinitiative"],
                 primary_knowledge_source=assigned_by,

src/monarch_ingest/ingests/go/annotation.yaml

@@ -83,6 +83,7 @@ edge_properties:
   - 'negated'
   - 'has_evidence'
   - 'object'
+  - 'publications'
   - 'aggregator_knowledge_source'
   - 'primary_knowledge_source'
   - 'knowledge_level'

tests/unit/go/test_go_annotation.py

@@ -504,5 +504,6 @@ def test_mgi_curie(mgi_entities):
     association = [association for association in mgi_entities if isinstance(association, Association)][0]
     assert association
     assert association.subject == "MGI:1918911"
+    assert association.publications == ["MGI:2156816", "GO_REF:0000015"]
     assert association.primary_knowledge_source == "infores:mgi"
     assert "infores:monarchinitiative" in association.aggregator_knowledge_source