Move copy & pasted oak similarity schema to it's own file

backend/src/monarch_py/datamodels/model.py

@@ -58,7 +58,7 @@ class Association(ConfiguredBaseModel):
         description="""Field containing subject id and the ids of all of it's ancestors""",
     )
     subject_label: Optional[str] = Field(
-        None, description="""the label or name for the first entity"""
+        None, description="""The name of the subject entity"""
     )
     subject_closure_label: Optional[List[str]] = Field(
         default_factory=list,
@@ -80,7 +80,7 @@ class Association(ConfiguredBaseModel):
         description="""Field containing object id and the ids of all of it's ancestors""",
     )
     object_label: Optional[str] = Field(
-        None, description="""the label or name for the second entity"""
+        None, description="""The name of the object entity"""
     )
     object_closure_label: Optional[List[str]] = Field(
         default_factory=list,
@@ -236,7 +236,7 @@ class DirectionalAssociation(Association):
         description="""Field containing subject id and the ids of all of it's ancestors""",
     )
     subject_label: Optional[str] = Field(
-        None, description="""the label or name for the first entity"""
+        None, description="""The name of the subject entity"""
     )
     subject_closure_label: Optional[List[str]] = Field(
         default_factory=list,
@@ -258,7 +258,7 @@ class DirectionalAssociation(Association):
         description="""Field containing object id and the ids of all of it's ancestors""",
     )
     object_label: Optional[str] = Field(
-        None, description="""the label or name for the second entity"""
+        None, description="""The name of the object entity"""
     )
     object_closure_label: Optional[List[str]] = Field(
         default_factory=list,
@@ -523,9 +523,7 @@ class SearchResult(Entity):
     highlight: Optional[str] = Field(
         None, description="""matching text snippet containing html tags"""
     )
-    score: Optional[str] = Field(
-        None, description="""Abstract base slot for different kinds of scores"""
-    )
+    score: Optional[float] = Field(None)
     id: str = Field(...)
     category: str = Field(...)
     name: str = Field(...)
@@ -582,7 +580,7 @@ class TermPairwiseSimilarity(PairwiseSimilarity):
         ..., description="""The first of the two entities being compared"""
     )
     subject_label: Optional[str] = Field(
-        None, description="""the label or name for the first entity"""
+        None, description="""The name of the subject entity"""
     )
     subject_source: Optional[str] = Field(
         None, description="""the source for the first entity"""
@@ -591,7 +589,7 @@ class TermPairwiseSimilarity(PairwiseSimilarity):
         None, description="""The second of the two entities being compared"""
     )
     object_label: Optional[str] = Field(
-        None, description="""the label or name for the second entity"""
+        None, description="""The name of the object entity"""
     )
     object_source: Optional[str] = Field(
         None, description="""the source for the second entity"""

backend/src/monarch_py/datamodels/model.yaml

@@ -4,10 +4,9 @@ description: Data models for the Monarch Initiative data access library
 prefixes:
   linkml: https://w3id.org/linkml/
   biolink: https://w3id.org/biolink/vocab/
-  oak: https://w3id.org/oak/
 imports:
   - linkml:types
-  - oak:similarity
+  - similarity
 default_range: string
 
 
@@ -239,90 +238,6 @@ classes:
         items:
           range: SearchResult
 
-# classes below a copy & paste from https://w3id.org/oak/similarity.yaml
-  PairwiseSimilarity:
-    abstract: true
-    description: >-
-      Abstract grouping for representing individual pairwise similarities
-
-  TermPairwiseSimilarity:
-    is_a: PairwiseSimilarity
-    description: >-
-      A simple pairwise similarity between two atomic concepts/terms
-    slots:
-      - subject_id
-      - subject_label
-      - subject_source
-      - object_id
-      - object_label
-      - object_source
-      - ancestor_id
-      - ancestor_label
-      - ancestor_source
-      - object_information_content
-      - subject_information_content
-      - ancestor_information_content
-      - jaccard_similarity
-      - cosine_similarity
-      - dice_similarity
-      - phenodigm_score
-
-  TermSetPairwiseSimilarity:
-    is_a: PairwiseSimilarity
-    description: >-
-      A simple pairwise similarity between two sets of concepts/terms
-    slots:
-      - subject_termset
-      - object_termset
-      - subject_best_matches
-      - object_best_matches
-      - average_score
-      - best_score
-      - metric
-
-  TermInfo:
-    attributes:
-      id:
-        identifier: true
-      label:
-        slot_uri: rdfs:label
-
-  BestMatch:
-    attributes:
-      match_source:
-        identifier: true
-        comments:
-          - note that the match_source is either the subject or the object
-      match_source_label:
-      match_target:
-        description: the entity matches
-      match_target_label:
-      score:
-        range: float
-        required: true
-      match_subsumer:
-        range: uriorcurie
-      match_subsumer_label:
-      similarity:
-        range: TermPairwiseSimilarity
-        required: true
-
-types:
-  # Types below are a copy & paste from https://w3id.org/oak/similarity.yaml
-  ZeroToOne:
-    typeof: float
-    minimum_value: 0
-    maximum_value: 0
-  NonNegativeFloat:
-    typeof: float
-    minimum_value: 0
-  NegativeLogValue:
-    typeof: float
-    minimum_value: 0
-  ItemCount:
-    typeof: integer
-    minimum_value: 0
-
 slots:
   aggregator_knowledge_source:
     multivalued: true
@@ -597,113 +512,3 @@ slots:
   stage_qualifier_closure_label:
     multivalued: true
     description: Field containing stage_qualifier name and the names of all of it's ancestors
-# slots below are a copy & paste from https://w3id.org/oak/similarity.yaml
-  subject_id:
-    slot_uri: sssom:subject_id
-    required: true
-    range: uriorcurie
-    description: The first of the two entities being compared
-# Excluded, since it conflicts with subject_label from this schema
-#  subject_label:
-#    slot_uri: sssom:subject_label
-#    description: the label or name for the first entity
-  subject_source:
-    slot_uri: sssom:subject_source
-    description: the source for the first entity
-  object_id:
-    slot_uri: sssom:object_id
-    range: uriorcurie
-    description: The second of the two entities being compared
-# Excluded, since it conflicts with object_label from this schema
-#  object_label:
-#    slot_uri: sssom:object_label
-#    description: the label or name for the second entity
-  object_source:
-    slot_uri: sssom:object_source
-    description: the source for the second entity
-  ancestor_id:
-    range: uriorcurie
-    description: >-
-      the most recent common ancestor of the two compared entities. If there are multiple MRCAs then
-      the most informative one is selected
-    todos:
-      - decide on what to do when there are multiple possible ancestos
-  ancestor_label:
-    description: the name or label of the ancestor concept
-  ancestor_source:
-# Excluded, conflicts with score from this schema
-#  score:
-#    abstract: true
-#    description: Abstract base slot for different kinds of scores
-  information_content:
-    abstract: true
-    aliases:
-      - IC
-    is_a: score
-    range: NegativeLogValue
-    description: The IC is the negative log of the probability of the concept
-  subject_information_content:
-    is_a: information_content
-    description: The IC of the subject
-  object_information_content:
-    is_a: information_content
-    description: The IC of the object
-  ancestor_information_content:
-    is_a: information_content
-    description: The IC of the object
-  jaccard_similarity:
-    is_a: score
-    range: ZeroToOne
-    description: The number of concepts in the intersection divided by the number in the union
-  cosine_similarity:
-     is_a: score
-     range: float
-     description: the dot product of two node embeddings divided by the product of their lengths
-  dice_similarity:
-    is_a: score
-    range: ZeroToOne
-  phenodigm_score:
-    is_a: score
-    range: NonNegativeFloat
-    description: the geometric mean of the jaccard similarity and the information content
-    equals_expression: sqrt({jaccard_similarity} * {information_content})
-  overlap_coefficient:
-    is_a: score
-    range: ZeroToOne
-  subsumes_score:
-    is_a: score
-    range: ZeroToOne
-  subsumed_by_score:
-    is_a: score
-    range: ZeroToOne
-  intersection_count:
-    is_a: score
-    range: ItemCount
-  union_count:
-    is_a: score
-    range: ItemCount
-  # TermSets
-  subject_termset:
-    range: TermInfo
-    multivalued: true
-    inlined: true
-  object_termset:
-    range: TermInfo
-    multivalued: true
-    inlined: true
-  subject_best_matches:
-    range: BestMatch
-    multivalued: true
-    inlined: true
-  object_best_matches:
-    range: BestMatch
-    multivalued: true
-    inlined: true
-  metric:
-    range: uriorcurie
-  average_score:
-    range: float
-    required: false
-  best_score:
-    range: float
-    required: false