feat!: update va-spec models to 1.0.0 (#28)

close #27 

* build: require stable versions of ga4gh.vrs + pydantic
* build: update vrs and cat-vrs dependencies

Author: korikuzma

.gitmodules

@@ -1,4 +1,4 @@
 [submodule "submodules/va_spec"]
 	path = submodules/va_spec
 	url = https://github.com/ga4gh/va-spec
-	branch = 1.0.0-ballot.2025-03
+	branch = 1.0

pyproject.toml

@@ -31,9 +31,9 @@ keywords = [
 requires-python = ">=3.10"
 dynamic = ["version"]
 dependencies = [
-    "ga4gh.vrs==2.*",
-    "ga4gh.cat_vrs~=0.6.0",
-    "pydantic==2.*"
+    "ga4gh.vrs>=2.1.3,<3.0",
+    "ga4gh.cat_vrs~=0.7.1",
+    "pydantic>=2.0,<3.0"
 ]
 
 [project.optional-dependencies]

src/ga4gh/va_spec/acmg_2015/models.py

@@ -38,9 +38,11 @@ class AcmgClassification(str, Enum):
 
 
 class VariantPathogenicityEvidenceLine(EvidenceLine):
-    """An Evidence Line that describes how information about the specific criterion
-    evidence for the variant was assessed as evidence for or against the variant's
-    pathogenicity.
+    """An Evidence Line that describes how a specific type of information was
+    interpreted as evidence for or againtst a variant's pathogenicity. In the ACMG
+    Framework, evidence is assessed by determining if a specific criterion (e.g. 'PM2')
+    with a default strength (e.g. 'moderate') is 'met' or 'not met', and in some cases
+    adjusting the default strength based on the quality and abundance of evidence.
     """
 
     targetProposition: VariantPathogenicityProposition | None = Field(
@@ -129,15 +131,15 @@ class VariantPathogenicityStatement(Statement):
 
     proposition: VariantPathogenicityProposition = Field(
         ...,
-        description="A proposition about the pathogenicity of a varaint, the validity of which is assessed and reported by the Statement. A Statement can put forth the proposition as being true, false, or uncertain, and may provide an assessment of the level of confidence/evidence supporting this claim.",
+        description="A proposition about the pathogenicity of a variant, the validity of which is assessed and reported by the Statement. A Statement can put forth the proposition as being true, false, or uncertain, and may provide an assessment of the level of confidence/evidence supporting this claim.",
     )
     strength: MappableConcept | None = Field(
         None,
         description="The strength of support that an ACMG 2015 Variant Pathogenicity statement is determined to provide for or against the proposed pathogenicity of the assessed variant. Strength is evaluated relative to the direction indicated by the 'direction' attribute. The indicated enumeration constrains the nested MappableConcept.primaryCoding > Coding.code attribute when capturing evidence strength.",
     )
     classification: MappableConcept = Field(
         ...,
-        description="The classification of the variant's pathogenicity, based on the ACMG 2015 guidelines. These classifications must coincide with the direction and strength values as follows: 'pathogenic' with supports-strong, 'likely pathogenic' with supports-moderate, 'benign' with disputes-strong, 'likely benign' with disputes-moderate 'uncertain significance' can be one of three possibilities... supports-weak, disputes-weak or neutral for uncertain significance (favoring pathogenic), uncertain significance (favoring benign) or uncertain significance (favoring neither pathogenic nor benign). The 'low penetrance' and 'risk allele' versions of pathogenicity classifications would be applied based on whether the variant proposition was defined to have a 'penetrance' of 'low' or 'risk' respectively.",
+        description="The classification of the variant's pathogenicity, based on the ACMG 2015 guidelines. These classifications should coincide with the direction and strength values as follows: 'pathogenic' with supports-strong, 'likely pathogenic' with supports-moderate, 'benign' with disputes-strong, 'likely benign' with disputes-moderate 'uncertain significance' can be one of three possibilities... supports-weak, disputes-weak or neutral for uncertain significance (favoring pathogenic), uncertain significance (favoring benign) or uncertain significance (favoring neither pathogenic nor benign). The 'low penetrance' and 'risk allele' versions of pathogenicity classifications would be applied based on whether the variant proposition was defined to have a 'penetrance' of 'low' or 'risk' respectively.",
     )
     specifiedBy: Method | iriReference = Field(
         ...,

src/ga4gh/va_spec/base/core.py

@@ -101,12 +101,27 @@ class Document(Entity, BaseModelForbidExtra):
         None,
         description="A [Digital Object Identifier](https://www.doi.org/the-identifier/what-is-a-doi/) for the document.",
     )
-    pmid: int | None = Field(
+    pmid: str | None = Field(
         None,
         description="A [PubMed unique identifier](https://en.wikipedia.org/wiki/PubMed#PubMed_identifier) for the document.",
     )
 
 
+class Method(Entity, BaseModelForbidExtra):
+    """A set of instructions that specify how to achieve some objective."""
+
+    type: Literal["Method"] = Field(
+        CoreType.METHOD.value, description=f"MUST be '{CoreType.METHOD.value}'."
+    )
+    methodType: str | None = Field(
+        None,
+        description="A specific type of method that a Method instance represents (e.g. 'Variant Interpretation Guideline', or 'Experimental Protocol').",
+    )
+    reportedIn: Document | iriReference | None = Field(
+        None, description="A document in which the the Method is reported."
+    )
+
+
 class InformationEntity(Entity):
     """An abstract (non-physical) entity that represents 'information content' carried by
     physical or digital information artifacts such as books, web pages, data sets, or
@@ -311,7 +326,7 @@ class ClinicalVariantProposition(_SubjectVariantPropositionBase):
     )
     alleleOriginQualifier: MappableConcept | iriReference | None = Field(
         None,
-        description="Reports whether the Proposition should be interpreted in the context of an inherited (germline) variant, an acquired (somatic) mutation, or another more nuanced concept. Consider using terms or codes from community terminologies here, e.g. terms from the 'allele origin' branch of the GENO ontology such as GENO:0000882 (somatic allele origin).",
+        description="Reports whether the Proposition should be interpreted in the context of a heritable 'germline' variant, an acquired 'somatic' variant in a tumor,  post-zygotic 'mosaic' variant. While these are the most commonly reported allele origins, other more nuanced concepts can be captured  (e.g. 'maternal' vs 'paternal' allele origin'). In practice, populating this field may be complicated by the fact that some sources report allele origin based on the type of tissue that was sequenced to identify the variant, and others use it more generally to specify a category of variant for which the proposition holds. The stated intent of this attribute is the latter. However, if an implementer is not sure about which is reported in their data, it may be safer to create an Extension to hold this information, where they can explicitly acknowledge this ambiguity.",
     )
 
 
@@ -326,9 +341,9 @@ class ExperimentalVariantFunctionalImpactProposition(
         "ExperimentalVariantFunctionalImpactProposition",
         description="MUST be 'ExperimentalVariantFunctionalImpactProposition'.",
     )
-    predicate: str = Field(
+    predicate: Literal["impactsFunctionOf"] = Field(
         "impactsFunctionOf",
-        description="The relationship the Proposition describes between the subject variant and object sequence feature whose function it may alter.",
+        description="The relationship the Proposition describes between the subject variant and object sequence feature whose function it may alter. MUST be 'impactsFunctionOf'.",
     )
     objectSequenceFeature: iriReference | MappableConcept = Field(
         ...,
@@ -351,7 +366,10 @@ class VariantDiagnosticProposition(ClinicalVariantProposition, BaseModelForbidEx
         "VariantDiagnosticProposition",
         description="MUST be 'VariantDiagnosticProposition'.",
     )
-    predicate: DiagnosticPredicate
+    predicate: DiagnosticPredicate = Field(
+        ...,
+        description="The relationship the Proposition describes between the subject variant and object Condition. MUST be one of 'isDiagnosticInclusionCriterionFor' or 'isDiagnosticExclusionCriterionFor'.",
+    )
     objectCondition: Condition | iriReference = Field(
         ..., description="The disease that is evaluated for diagnosis."
     )
@@ -364,7 +382,10 @@ class VariantOncogenicityProposition(ClinicalVariantProposition, BaseModelForbid
         "VariantOncogenicityProposition",
         description="MUST be 'VariantOncogenicityProposition'.",
     )
-    predicate: str = "isCausalFor"
+    predicate: Literal["isOncogenicFor"] = Field(
+        "isOncogenicFor",
+        description="The relationship the Proposition describes between the subject variant and object tumor type. MUST be 'isOncogenicFor'.",
+    )
     objectTumorType: Condition | iriReference = Field(
         ..., description="The tumor type for which the variant impact is evaluated."
     )
@@ -377,7 +398,10 @@ class VariantPathogenicityProposition(ClinicalVariantProposition, BaseModelForbi
         "VariantPathogenicityProposition",
         description="Must be 'VariantPathogenicityProposition'",
     )
-    predicate: str = "isCausalFor"
+    predicate: Literal["isCausalFor"] = Field(
+        "isCausalFor",
+        description="The relationship the Proposition describes between the subject variant and object condition. MUST be 'isCausalFor'.",
+    )
     objectCondition: Condition | iriReference = Field(
         ..., description="The Condition for which the variant impact is stated."
     )
@@ -400,7 +424,10 @@ class VariantPrognosticProposition(ClinicalVariantProposition, BaseModelForbidEx
         "VariantPrognosticProposition",
         description="MUST be 'VariantPrognosticProposition'.",
     )
-    predicate: PrognosticPredicate
+    predicate: PrognosticPredicate = Field(
+        ...,
+        description="The relationship the Proposition describes between the subject variant and object Condition. MUST be one of 'associatedWithBetterOutcomeFor' or 'associatedWithWorseOutcomeFor'.",
+    )
     objectCondition: Condition | iriReference = Field(
         ..., description="The disease that is evaluated for outcome."
     )
@@ -419,7 +446,10 @@ class VariantTherapeuticResponseProposition(
         "VariantTherapeuticResponseProposition",
         description="MUST be 'VariantTherapeuticResponseProposition'.",
     )
-    predicate: TherapeuticResponsePredicate
+    predicate: TherapeuticResponsePredicate = Field(
+        ...,
+        description="The relationship the Proposition describes between the subject variant and object theapeutic. MUST be one of 'predictsSensitivityTo' or 'predictsResistanceTo'.",
+    )
     objectTherapeutic: Therapeutic | iriReference = Field(
         ...,
         description="A drug administration or other therapeutic procedure that the neoplasm is intended to respond to.",
@@ -430,21 +460,6 @@ class VariantTherapeuticResponseProposition(
     )
 
 
-class Method(Entity, BaseModelForbidExtra):
-    """A set of instructions that specify how to achieve some objective."""
-
-    type: Literal["Method"] = Field(
-        CoreType.METHOD.value, description=f"MUST be '{CoreType.METHOD.value}'."
-    )
-    methodType: str | None = Field(
-        None,
-        description="A specific type of method that a Method instance represents (e.g. 'Variant Interpretation Guideline', or 'Experimental Protocol').",
-    )
-    reportedIn: Document | iriReference | None = Field(
-        None, description="A document in which the the Method is reported."
-    )
-
-
 class Agent(Entity, BaseModelForbidExtra):
     """An autonomous actor (person, organization, or software agent) that bears some
     form of responsibility for an activity taking place, for the existence of an entity,

submodules/va_spec

@@ -169,7 +169,7 @@ def test_evidence_line(caf):
                         "name": "Danos et al., 2019, Genome Med.",
                         "title": "Standard operating procedure for curation and clinical interpretation of variants in cancer",
                         "doi": "10.1186/s13073-019-0687-x",
-                        "pmid": 31779674,
+                        "pmid": "31779674",
                         "type": "Document",
                     },
                     "type": "Method",
@@ -254,7 +254,7 @@ def test_variant_pathogenicity_stmt():
         "specifiedBy": {
             "reportedIn": {
                 "type": "Document",
-                "pmid": 25741868,
+                "pmid": "25741868",
                 "name": "ACMG Guidelines, 2015",
             }
         },
@@ -305,7 +305,7 @@ def test_variant_pathogenicity_el():
             "name": "ACMG 2015 PS3 Criterion",
             "reportedIn": {
                 "type": "Document",
-                "pmid": 25741868,
+                "pmid": "25741868",
                 "name": "ACMG Guidelines, 2015",
             },
             "methodType": "PS3",
@@ -413,7 +413,7 @@ def test_variant_onco_stmt():
         "direction": "neutral",
         "proposition": {
             "type": "VariantOncogenicityProposition",
-            "predicate": "isCausalFor",
+            "predicate": "isOncogenicFor",
             "objectTumorType": "conditions.json#/1",
             "subjectVariant": "alleles.json#/1",
         },
@@ -468,7 +468,7 @@ def test_variant_onco_el():
             "type": "Method",
             "reportedIn": {
                 "type": "Document",
-                "pmid": 35101336,
+                "pmid": "35101336",
                 "name": "ClinGen/CGC/VICC Guidelines for Oncogenicity, 2022",
             },
             "methodType": "OS2",

tests/validation/test_va_spec_models.py

@@ -28,7 +28,7 @@ class VaSpecSchemaMapping(BaseModel):
     base_classes: set = set()
     concrete_classes: set = set()
     primitives: set = set()
-    schema: dict = {}
+    va_spec_schema: dict = {}
 
 
 def _update_va_spec_schema_mapping(
@@ -43,7 +43,7 @@ def _update_va_spec_schema_mapping(
         cls_def = json.load(rf)
 
     spec_class = cls_def["title"]
-    va_spec_schema_mapping.schema[spec_class] = cls_def
+    va_spec_schema_mapping.va_spec_schema[spec_class] = cls_def
 
     if "properties" in cls_def:
         va_spec_schema_mapping.concrete_classes.add(spec_class)
@@ -108,11 +108,11 @@ def test_schema_class_fields(va_spec_schema, pydantic_models):
     """
     mapping = VA_SPEC_SCHEMA_MAPPING[va_spec_schema]
     for schema_model in mapping.concrete_classes:
-        schema_properties = mapping.schema[schema_model]["properties"]
+        schema_properties = mapping.va_spec_schema[schema_model]["properties"]
         pydantic_model = getattr(pydantic_models, schema_model)
         assert set(pydantic_model.model_fields) == set(schema_properties), schema_model
 
-        required_schema_fields = set(mapping.schema[schema_model]["required"])
+        required_schema_fields = set(mapping.va_spec_schema[schema_model]["required"])
 
         for prop, property_def in schema_properties.items():
             pydantic_model_field_info = pydantic_model.model_fields[prop]