Add medical action association table to UI (plus fixture & enum regen)

.github/workflows/test-backend.yaml

@@ -4,6 +4,9 @@ on:
   pull_request:
   workflow_dispatch:
 
+env:
+  PYTHONPATH: ${{ github.workspace }}/backend
+
 # defaults:
 #   run:
 #     working-directory: ./backend
@@ -40,14 +43,16 @@ jobs:
       #              run pytest
       #----------------------------------------------
       - name: Run tests
-        run: poetry -C backend run pytest backend/tests
+        run: |
+          poetry -C backend run pytest tests
+        shell: bash
 
       #----------------------------------------------
       #              coverage report
       #----------------------------------------------
       - name: Generate coverage results
         run: |
-          poetry -C backend run coverage run -m pytest backend/tests
+          poetry -C backend run coverage run -m pytest tests
           poetry -C backend run coverage xml
           poetry -C backend run coverage report -m
         shell: bash

backend/src/monarch_py/datamodels/category_enums.py

@@ -4,119 +4,60 @@
 class EntityCategory(Enum):
     """Entity categories"""
 
-    SEQUENCE_VARIANT = "biolink:SequenceVariant"
     GENE = "biolink:Gene"
-    GENOTYPE = "biolink:Genotype"
     PHENOTYPIC_FEATURE = "biolink:PhenotypicFeature"
-    BIOLOGICAL_PROCESS_OR_ACTIVITY = "biolink:BiologicalProcessOrActivity"
-    DISEASE = "biolink:Disease"
-    GROSS_ANATOMICAL_STRUCTURE = "biolink:GrossAnatomicalStructure"
-    CELL = "biolink:Cell"
-    PATHWAY = "biolink:Pathway"
-    NAMED_THING = "biolink:NamedThing"
+    GENOTYPE = "biolink:Genotype"
     ANATOMICAL_ENTITY = "biolink:AnatomicalEntity"
-    CELLULAR_COMPONENT = "biolink:CellularComponent"
-    MOLECULAR_ENTITY = "biolink:MolecularEntity"
     BIOLOGICAL_PROCESS = "biolink:BiologicalProcess"
-    MACROMOLECULAR_COMPLEX = "biolink:MacromolecularComplex"
-    MOLECULAR_ACTIVITY = "biolink:MolecularActivity"
+    DISEASE = "biolink:Disease"
+    NAMED_THING = "biolink:NamedThing"
+    PATHWAY = "biolink:Pathway"
     PROTEIN = "biolink:Protein"
-    CELLULAR_ORGANISM = "biolink:CellularOrganism"
-    VERTEBRATE = "biolink:Vertebrate"
-    VIRUS = "biolink:Virus"
-    BEHAVIORAL_FEATURE = "biolink:BehavioralFeature"
     CHEMICAL_ENTITY = "biolink:ChemicalEntity"
-    LIFE_STAGE = "biolink:LifeStage"
-    PATHOLOGICAL_PROCESS = "biolink:PathologicalProcess"
-    DRUG = "biolink:Drug"
-    SMALL_MOLECULE = "biolink:SmallMolecule"
+    SEQUENCE_VARIANT = "biolink:SequenceVariant"
+    MOLECULAR_ACTIVITY = "biolink:MolecularActivity"
+    CELLULAR_COMPONENT = "biolink:CellularComponent"
+    CELL = "biolink:Cell"
     ORGANISM_TAXON = "biolink:OrganismTaxon"
-    INFORMATION_CONTENT_ENTITY = "biolink:InformationContentEntity"
-    NUCLEIC_ACID_ENTITY = "biolink:NucleicAcidEntity"
-    EVIDENCE_TYPE = "biolink:EvidenceType"
-    RNAPRODUCT = "biolink:RNAProduct"
-    TRANSCRIPT = "biolink:Transcript"
-    FUNGUS = "biolink:Fungus"
-    PLANT = "biolink:Plant"
-    PROCESSED_MATERIAL = "biolink:ProcessedMaterial"
-    ACTIVITY = "biolink:Activity"
-    AGENT = "biolink:Agent"
-    CONFIDENCE_LEVEL = "biolink:ConfidenceLevel"
-    DATASET = "biolink:Dataset"
-    ENVIRONMENTAL_FEATURE = "biolink:EnvironmentalFeature"
-    GENETIC_INHERITANCE = "biolink:GeneticInheritance"
-    HAPLOTYPE = "biolink:Haplotype"
-    INVERTEBRATE = "biolink:Invertebrate"
-    MAMMAL = "biolink:Mammal"
-    POPULATION_OF_INDIVIDUAL_ORGANISMS = "biolink:PopulationOfIndividualOrganisms"
-    PROTEIN_FAMILY = "biolink:ProteinFamily"
-    PUBLICATION = "biolink:Publication"
-    ACCESSIBLE_DNA_REGION = "biolink:AccessibleDnaRegion"
-    BACTERIUM = "biolink:Bacterium"
-    BIOLOGICAL_SEX = "biolink:BiologicalSex"
-    CELL_LINE = "biolink:CellLine"
-    CHEMICAL_EXPOSURE = "biolink:ChemicalExposure"
-    CHEMICAL_MIXTURE = "biolink:ChemicalMixture"
-    DATASET_DISTRIBUTION = "biolink:DatasetDistribution"
-    DIAGNOSTIC_AID = "biolink:DiagnosticAid"
-    DRUG_EXPOSURE = "biolink:DrugExposure"
-    ENVIRONMENTAL_PROCESS = "biolink:EnvironmentalProcess"
-    EVENT = "biolink:Event"
-    EXON = "biolink:Exon"
-    GENOME = "biolink:Genome"
-    GENOTYPIC_SEX = "biolink:GenotypicSex"
-    HUMAN = "biolink:Human"
-    INDIVIDUAL_ORGANISM = "biolink:IndividualOrganism"
-    MATERIAL_SAMPLE = "biolink:MaterialSample"
-    MICRO_RNA = "biolink:MicroRNA"
-    ORGANISMAL_ENTITY = "biolink:OrganismalEntity"
-    PATENT = "biolink:Patent"
-    PHENOTYPIC_SEX = "biolink:PhenotypicSex"
-    POLYPEPTIDE = "biolink:Polypeptide"
-    PROTEIN_DOMAIN = "biolink:ProteinDomain"
-    REAGENT_TARGETED_GENE = "biolink:ReagentTargetedGene"
-    REGULATORY_REGION = "biolink:RegulatoryRegion"
-    SI_RNA = "biolink:SiRNA"
-    SNV = "biolink:Snv"
-    STUDY = "biolink:Study"
-    STUDY_VARIABLE = "biolink:StudyVariable"
-    TRANSCRIPTION_FACTOR_BINDING_SITE = "biolink:TranscriptionFactorBindingSite"
-    TREATMENT = "biolink:Treatment"
-    WEB_PAGE = "biolink:WebPage"
-    ZYGOSITY = "biolink:Zygosity"
+    MOLECULAR_ENTITY = "biolink:MolecularEntity"
+    LIFE_STAGE = "biolink:LifeStage"
 
 
 class AssociationCategory(Enum):
     """Association categories"""
 
     PAIRWISE_GENE_TO_GENE_INTERACTION = "biolink:PairwiseGeneToGeneInteraction"
     GENE_TO_EXPRESSION_SITE_ASSOCIATION = "biolink:GeneToExpressionSiteAssociation"
-    VARIANT_TO_GENE_ASSOCIATION = "biolink:VariantToGeneAssociation"
     MACROMOLECULAR_MACHINE_TO_BIOLOGICAL_PROCESS_ASSOCIATION = (
         "biolink:MacromolecularMachineToBiologicalProcessAssociation"
     )
     GENE_TO_PHENOTYPIC_FEATURE_ASSOCIATION = "biolink:GeneToPhenotypicFeatureAssociation"
+    ASSOCIATION = "biolink:Association"
     MACROMOLECULAR_MACHINE_TO_MOLECULAR_ACTIVITY_ASSOCIATION = (
         "biolink:MacromolecularMachineToMolecularActivityAssociation"
     )
     MACROMOLECULAR_MACHINE_TO_CELLULAR_COMPONENT_ASSOCIATION = (
         "biolink:MacromolecularMachineToCellularComponentAssociation"
     )
-    ASSOCIATION = "biolink:Association"
     GENE_TO_GENE_HOMOLOGY_ASSOCIATION = "biolink:GeneToGeneHomologyAssociation"
     GENOTYPE_TO_PHENOTYPIC_FEATURE_ASSOCIATION = "biolink:GenotypeToPhenotypicFeatureAssociation"
     DISEASE_TO_PHENOTYPIC_FEATURE_ASSOCIATION = "biolink:DiseaseToPhenotypicFeatureAssociation"
     GENE_TO_PATHWAY_ASSOCIATION = "biolink:GeneToPathwayAssociation"
-    DISEASE_OR_PHENOTYPIC_FEATURE_TO_LOCATION_ASSOCIATION = "biolink:DiseaseOrPhenotypicFeatureToLocationAssociation"
     CHEMICAL_TO_PATHWAY_ASSOCIATION = "biolink:ChemicalToPathwayAssociation"
+    VARIANT_TO_GENE_ASSOCIATION = "biolink:VariantToGeneAssociation"
+    VARIANT_TO_DISEASE_ASSOCIATION = "biolink:VariantToDiseaseAssociation"
     GENOTYPE_TO_DISEASE_ASSOCIATION = "biolink:GenotypeToDiseaseAssociation"
     CORRELATED_GENE_TO_DISEASE_ASSOCIATION = "biolink:CorrelatedGeneToDiseaseAssociation"
+    CHEMICAL_TO_DISEASE_OR_PHENOTYPIC_FEATURE_ASSOCIATION = "biolink:ChemicalToDiseaseOrPhenotypicFeatureAssociation"
     DISEASE_OR_PHENOTYPIC_FEATURE_TO_GENETIC_INHERITANCE_ASSOCIATION = (
         "biolink:DiseaseOrPhenotypicFeatureToGeneticInheritanceAssociation"
     )
-    VARIANT_TO_DISEASE_ASSOCIATION = "biolink:VariantToDiseaseAssociation"
     CAUSAL_GENE_TO_DISEASE_ASSOCIATION = "biolink:CausalGeneToDiseaseAssociation"
-    CHEMICAL_TO_DISEASE_OR_PHENOTYPIC_FEATURE_ASSOCIATION = "biolink:ChemicalToDiseaseOrPhenotypicFeatureAssociation"
+    DISEASE_OR_PHENOTYPIC_FEATURE_TO_LOCATION_ASSOCIATION = "biolink:DiseaseOrPhenotypicFeatureToLocationAssociation"
+    CHEMICAL_OR_DRUG_OR_TREATMENT_TO_DISEASE_OR_PHENOTYPIC_FEATURE_ASSOCIATION = (
+        "biolink:ChemicalOrDrugOrTreatmentToDiseaseOrPhenotypicFeatureAssociation"
+    )
+    VARIANT_TO_PHENOTYPIC_FEATURE_ASSOCIATION = "biolink:VariantToPhenotypicFeatureAssociation"
 
 
 class AssociationPredicate(Enum):
@@ -125,33 +66,40 @@ class AssociationPredicate(Enum):
     INTERACTS_WITH = "biolink:interacts_with"
     EXPRESSED_IN = "biolink:expressed_in"
     HAS_PHENOTYPE = "biolink:has_phenotype"
-    IS_SEQUENCE_VARIANT_OF = "biolink:is_sequence_variant_of"
     ENABLES = "biolink:enables"
     ACTIVELY_INVOLVED_IN = "biolink:actively_involved_in"
     ORTHOLOGOUS_TO = "biolink:orthologous_to"
-    LOCATED_IN = "biolink:located_in"
     SUBCLASS_OF = "biolink:subclass_of"
+    LOCATED_IN = "biolink:located_in"
     RELATED_TO = "biolink:related_to"
     PARTICIPATES_IN = "biolink:participates_in"
     ACTS_UPSTREAM_OF_OR_WITHIN = "biolink:acts_upstream_of_or_within"
-    ACTIVE_IN = "biolink:active_in"
+    IS_ACTIVE_IN = "biolink:is_active_in"
     PART_OF = "biolink:part_of"
-    MODEL_OF = "biolink:model_of"
     CAUSES = "biolink:causes"
+    IS_SEQUENCE_VARIANT_OF = "biolink:is_sequence_variant_of"
+    MODEL_OF = "biolink:model_of"
     ACTS_UPSTREAM_OF = "biolink:acts_upstream_of"
+    TREATS_OR_APPLIED_OR_STUDIED_TO_TREAT = "biolink:treats_or_applied_or_studied_to_treat"
     HAS_MODE_OF_INHERITANCE = "biolink:has_mode_of_inheritance"
-    CONTRIBUTES_TO = "biolink:contributes_to"
     GENE_ASSOCIATED_WITH_CONDITION = "biolink:gene_associated_with_condition"
-    TREATS_OR_APPLIED_OR_STUDIED_TO_TREAT = "biolink:treats_or_applied_or_studied_to_treat"
+    CONTRIBUTES_TO = "biolink:contributes_to"
+    ASSOCIATED_WITH_INCREASED_LIKELIHOOD_OF = "biolink:associated_with_increased_likelihood_of"
     COLOCALIZES_WITH = "biolink:colocalizes_with"
+    GENETICALLY_ASSOCIATED_WITH = "biolink:genetically_associated_with"
+    DISEASE_HAS_LOCATION = "biolink:disease_has_location"
     ACTS_UPSTREAM_OF_POSITIVE_EFFECT = "biolink:acts_upstream_of_positive_effect"
     ACTS_UPSTREAM_OF_OR_WITHIN_POSITIVE_EFFECT = "biolink:acts_upstream_of_or_within_positive_effect"
+    AMELIORATES_CONDITION = "biolink:ameliorates_condition"
     ACTS_UPSTREAM_OF_NEGATIVE_EFFECT = "biolink:acts_upstream_of_negative_effect"
     ACTS_UPSTREAM_OF_OR_WITHIN_NEGATIVE_EFFECT = "biolink:acts_upstream_of_or_within_negative_effect"
+    PREVENTATIVE_FOR_CONDITION = "biolink:preventative_for_condition"
+    CONTRAINDICATED_IN = "biolink:contraindicated_in"
 
 
 class MappingPredicate(Enum):
     """Mapping predicates"""
 
     EXACT_MATCH = "skos:exactMatch"
+    CLOSE_MATCH = "skos:closeMatch"
     BROAD_MATCH = "skos:broadMatch"

backend/src/monarch_py/implementations/solr/solr_parsers.py

@@ -91,7 +91,7 @@
 
 def parse_association_counts(query_result: SolrQueryResult, entity: str) -> AssociationCountList:
     subject_query = f'AND (subject:"{entity}" OR subject_closure:"{entity}")'
-    object_query = f'AND (object:"{entity}" OR object_closure:"{entity}")'
+    object_query = f'AND (object:"{entity}" OR object_closure:"{entity}" OR disease_context_qualifier:"{entity}" OR disease_context_qualifier_closure:"{entity}")'
     association_count_dict: Dict[str, AssociationCount] = {}
     for k, v in query_result.facet_counts.facet_queries.items():
         if v > 0:
@@ -258,6 +258,13 @@
         document.get("object_closure") and any(e in document.get("object_closure") for e in entity)
     ):
         direction = AssociationDirectionEnum.incoming
+    elif document.get("disease_context_qualifier") in entity or (
+        document.get("disease_context_qualifier_closure")
+        and any(e in document.get("disease_context_qualifier_closure") for e in entity)
+    ):
+        # This is a special case for disease_context_qualifier, if an association between two other entities
+        # only occurs within the context of a disease, we can treat it like an incoming association
+        direction = AssociationDirectionEnum.incoming
     else:
         raise ValueError(f"Entity {entity} not found in association {document}")
     return direction

backend/src/monarch_py/implementations/solr/solr_query_utils.py

@@ -30,6 +30,8 @@ def build_association_query(
     offset: int = 0,
     limit: int = 20,
 ) -> SolrQuery:
+
+    entity_fields = ["subject", "object", "disease_context_qualifier"]
     """Populate a SolrQuery object with association filters"""
     query = SolrQuery(start=offset, rows=limit)
     query.add_field_filter_query("category", None if not category else [c for c in category])
@@ -54,12 +56,10 @@ def build_association_query(
             query.add_filter_query(" OR ".join([f'object:"{o}" OR object_closure:"{o}"' for o in object]))
     if entity:
         if direct:
-            query.add_filter_query(" OR ".join([f'subject:"{e}" OR object:"{e}"' for e in entity]))
+            query.add_filter_query(" OR ".join([f'{field}:"{e}"' for e in entity for field in entity_fields]))
         else:
             query.add_filter_query(
-                " OR ".join(
-                    [f'subject:"{e}" OR subject_closure:"{e}" OR object:"{e}" OR object_closure:"{e}"' for e in entity]
-                )
+                " OR ".join([f'{field}:"{e}" OR {field}_closure:"{e}"' for e in entity for field in entity_fields])
             )
     if q:
         # We don't yet have tokenization strategies for the association index, initially we'll limit searching to
@@ -118,7 +118,8 @@ def build_association_table_query(
 
 def build_association_counts_query(entity: str) -> SolrQuery:
     subject_query = f'AND (subject:"{entity}" OR subject_closure:"{entity}")'
-    object_query = f'AND (object:"{entity}" OR object_closure:"{entity}")'
+    object_query = f'AND (object:"{entity}" OR object_closure:"{entity}" OR disease_context_qualifier:"{entity}" OR disease_context_qualifier_closure:"{entity}")'
+
     # Run the same facet_queries constrained to matches against either the subject or object
     # to know which kind of label will be needed in the UI to refer to the opposite side of the association
     facet_queries = []

backend/src/monarch_py/implementations/spacy/spacy_implementation.py

@@ -10,6 +10,7 @@
 import pystow
 import tarfile
 
+
 @dataclass
 class SpacyImplementation(TextAnnotatorInterface):
     """Implementation of Monarch Interfaces for SPACY"""
@@ -32,22 +33,31 @@ def init_spacy(self, grounding_implementation: GroundingInterface):
             with tarfile.open(model_archive, "r:gz") as tar:
                 tar.extractall(path=model_dir.parent)
 
-        model_subdir = next((d for d in model_archive.parent.iterdir() if d.is_dir() and d.name.startswith("en_core_sci")), None)
+        model_subdir = next(
+            (d for d in model_archive.parent.iterdir() if d.is_dir() and d.name.startswith("en_core_sci")), None
+        )
 
         if model_subdir:
-            inner_model_dir = next((d for d in model_subdir.iterdir() if d.is_dir() and d.name.startswith("en_core_sci") and "egg-info" not in d.name), None)
+            inner_model_dir = next(
+                (
+                    d
+                    for d in model_subdir.iterdir()
+                    if d.is_dir() and d.name.startswith("en_core_sci") and "egg-info" not in d.name
+                ),
+                None,
+            )
             if inner_model_dir:
                 # Load the model
-                self.nlp = spacy.load(str(str(model_archive.parent / model_subdir.name / inner_model_dir.name / model_subdir.name)))
+                self.nlp = spacy.load(
+                    str(str(model_archive.parent / model_subdir.name / inner_model_dir.name / model_subdir.name))
+                )
 
         # Assign the grounding implementation
         self.grounding_implementation = grounding_implementation
 
         # Test the model with a sample sentence
         self.nlp("Nystagmus, strabismus, fundus, ocular albinism, lewis.")
 
-
-
     def get_annotated_entities(self, text) -> List[TextAnnotationResult]:
         """Annotate text using SPACY"""
         results: List[TextAnnotationResult] = []

backend/src/monarch_py/utils/association_type_mappings.yaml

@@ -47,4 +47,7 @@
   category: "biolink:GenotypeToPhenotypicFeatureAssociation"
 - subject_label: Disease Model
   object_label: Disease Model
-  category: "biolink:GenotypeToDiseaseAssociation"
+  category: "biolink:GenotypeToDiseaseAssociation"
+- subject_label: Medical Action
+  object_label: Medical Action
+  category: "biolink:ChemicalOrDrugOrTreatmentToDiseaseOrPhenotypicFeatureAssociation"

backend/tests/fixtures/association_counts.py

@@ -7,12 +7,18 @@ def association_counts():
         "items": [
             {
                 "label": "Disease to Phenotype",
-                "count": 4082,
+                "count": 4077,
                 "category": "biolink:DiseaseToPhenotypicFeatureAssociation",
             },
+            {"label": "Gene to Phenotype", "count": 6350, "category": "biolink:GeneToPhenotypicFeatureAssociation"},
             {"label": "Causal Gene", "count": 125, "category": "biolink:CausalGeneToDiseaseAssociation"},
             {"label": "Correlated Gene", "count": 150, "category": "biolink:CorrelatedGeneToDiseaseAssociation"},
             {"label": "Variant to Disease", "count": 1, "category": "biolink:VariantToDiseaseAssociation"},
             {"label": "Disease Model", "count": 238, "category": "biolink:GenotypeToDiseaseAssociation"},
+            {
+                "label": "Medical Action",
+                "count": 4,
+                "category": "biolink:ChemicalOrDrugOrTreatmentToDiseaseOrPhenotypicFeatureAssociation",
+            },
         ]
     }