rdf author fields serialisation fix

src/cell_annotation_schema/curie_utils.py

@@ -0,0 +1,39 @@
+import os
+import json
+from pathlib import Path
+
+from importlib import resources
+from cell_annotation_schema import resource
+
+
+class CurieToIriConverter:
+
+    DEFAULT_JSONLD_FILE_NAME = 'prefixes.jsonld'
+
+    def __init__(self, jsonld_file=DEFAULT_JSONLD_FILE_NAME):
+        self.jsonld_file = jsonld_file
+        self.prefix_mappings = self.load_prefix_mappings(self.jsonld_file)
+
+    @staticmethod
+    def load_prefix_mappings(jsonld_file_name):
+        prefix_resource_file = resources.files(resource) / jsonld_file_name
+        if os.path.exists(jsonld_file_name):
+            # read from the path provided by the user
+            with open(jsonld_file_name, 'r') as file:
+                prefixes = json.load(file)
+        elif os.path.exists(prefix_resource_file):
+            # read from importlib.resources (resource) package
+            prefixes = json.loads(Path(prefix_resource_file).read_text())
+        else:
+            # read from resource folder
+            resource_dir = os.path.join(os.path.dirname(__file__), "./resource")
+            prefixes_path = os.path.join(resource_dir, jsonld_file_name)
+            prefixes = json.loads(Path(prefixes_path).read_text())
+        return prefixes.get('@context', {})
+
+    def curie_to_iri(self, curie):
+        prefix, reference = curie.split(':', 1)
+        if prefix in self.prefix_mappings:
+            return self.prefix_mappings[prefix] + reference
+        else:
+            raise ValueError(f"Prefix '{prefix}' not found in prefix mappings.")

src/cell_annotation_schema/datamodel/bican/cell_annotation_schema.py

@@ -1,5 +1,5 @@
 # Auto generated from None by pythongen.py version: 0.0.1
-# Generation date: 2024-09-23T16:32:47
+# Generation date: 2024-10-03T17:13:03
 # Schema: General_Cell_Annotation_Open_Standard
 #
 # id: https://cellular-semantics.sanger.ac.uk/ontology/CAS
@@ -33,6 +33,7 @@
 # Namespaces
 CAS = CurieNamespace('CAS', 'https://cellular-semantics.sanger.ac.uk/ontology/CAS/')
 CL = CurieNamespace('CL', 'http://purl.obolibrary.org/obo/CL_')
+CELLXGENE_DATASET = CurieNamespace('CellXGene_dataset', 'https://cellxgene.cziscience.com/datasets/')
 IAO = CurieNamespace('IAO', 'http://purl.obolibrary.org/obo/IAO_')
 PCL = CurieNamespace('PCL', 'http://purl.obolibrary.org/obo/PCL_')
 RO = CurieNamespace('RO', 'http://purl.obolibrary.org/obo/RO_')
@@ -513,7 +514,7 @@ class Taxonomy(YAMLRoot):
     title: str = None
     author_name: str = None
     labelsets: Union[Union[dict, Labelset], List[Union[dict, Labelset]]] = None
-    matrix_file_id: Optional[str] = None
+    matrix_file_id: Optional[Union[str, URIorCURIE]] = None
     description: Optional[str] = None
     cellannotation_schema_version: Optional[str] = None
     cellannotation_timestamp: Optional[str] = None
@@ -540,8 +541,8 @@ def __post_init__(self, *_: List[str], **kwargs: Dict[str, Any]):
             self.labelsets = [self.labelsets] if self.labelsets is not None else []
         self.labelsets = [v if isinstance(v, Labelset) else Labelset(**as_dict(v)) for v in self.labelsets]
 
-        if self.matrix_file_id is not None and not isinstance(self.matrix_file_id, str):
-            self.matrix_file_id = str(self.matrix_file_id)
+        if self.matrix_file_id is not None and not isinstance(self.matrix_file_id, URIorCURIE):
+            self.matrix_file_id = URIorCURIE(self.matrix_file_id)
 
         if self.description is not None and not isinstance(self.description, str):
             self.description = str(self.description)
@@ -14356,7 +14357,7 @@ class slots:
                    model_uri=CELL_ANNOTATION_SCHEMA.author_annotation_fields, domain=None, range=Optional[Union[dict, Any]])
 
 slots.matrix_file_id = Slot(uri=CELL_ANNOTATION_SCHEMA.matrix_file_id, name="matrix_file_id", curie=CELL_ANNOTATION_SCHEMA.curie('matrix_file_id'),
-                   model_uri=CELL_ANNOTATION_SCHEMA.matrix_file_id, domain=None, range=Optional[str])
+                   model_uri=CELL_ANNOTATION_SCHEMA.matrix_file_id, domain=None, range=Optional[Union[str, URIorCURIE]])
 
 slots.title = Slot(uri=CELL_ANNOTATION_SCHEMA.title, name="title", curie=CELL_ANNOTATION_SCHEMA.curie('title'),
                    model_uri=CELL_ANNOTATION_SCHEMA.title, domain=None, range=str)

src/cell_annotation_schema/datamodel/cap/cell_annotation_schema.py

@@ -1,5 +1,5 @@
 # Auto generated from None by pythongen.py version: 0.0.1
-# Generation date: 2024-09-23T16:32:48
+# Generation date: 2024-10-03T17:13:03
 # Schema: General_Cell_Annotation_Open_Standard
 #
 # id: https://cellular-semantics.sanger.ac.uk/ontology/CAS
@@ -33,6 +33,7 @@
 # Namespaces
 CAS = CurieNamespace('CAS', 'https://cellular-semantics.sanger.ac.uk/ontology/CAS/')
 CL = CurieNamespace('CL', 'http://purl.obolibrary.org/obo/CL_')
+CELLXGENE_DATASET = CurieNamespace('CellXGene_dataset', 'https://cellxgene.cziscience.com/datasets/')
 IAO = CurieNamespace('IAO', 'http://purl.obolibrary.org/obo/IAO_')
 PCL = CurieNamespace('PCL', 'http://purl.obolibrary.org/obo/PCL_')
 RO = CurieNamespace('RO', 'http://purl.obolibrary.org/obo/RO_')
@@ -440,7 +441,7 @@ class Taxonomy(YAMLRoot):
     title: str = None
     author_name: str = None
     labelsets: Union[Union[dict, Labelset], List[Union[dict, Labelset]]] = None
-    matrix_file_id: Optional[str] = None
+    matrix_file_id: Optional[Union[str, URIorCURIE]] = None
     description: Optional[str] = None
     cellannotation_schema_version: Optional[str] = None
     cellannotation_timestamp: Optional[str] = None
@@ -467,8 +468,8 @@ def __post_init__(self, *_: List[str], **kwargs: Dict[str, Any]):
             self.labelsets = [self.labelsets] if self.labelsets is not None else []
         self.labelsets = [v if isinstance(v, Labelset) else Labelset(**as_dict(v)) for v in self.labelsets]
 
-        if self.matrix_file_id is not None and not isinstance(self.matrix_file_id, str):
-            self.matrix_file_id = str(self.matrix_file_id)
+        if self.matrix_file_id is not None and not isinstance(self.matrix_file_id, URIorCURIE):
+            self.matrix_file_id = URIorCURIE(self.matrix_file_id)
 
         if self.description is not None and not isinstance(self.description, str):
             self.description = str(self.description)
@@ -14325,7 +14326,7 @@ class slots:
                    model_uri=CELL_ANNOTATION_SCHEMA.author_annotation_fields, domain=None, range=Optional[Union[dict, Any]])
 
 slots.matrix_file_id = Slot(uri=CELL_ANNOTATION_SCHEMA.matrix_file_id, name="matrix_file_id", curie=CELL_ANNOTATION_SCHEMA.curie('matrix_file_id'),
-                   model_uri=CELL_ANNOTATION_SCHEMA.matrix_file_id, domain=None, range=Optional[str])
+                   model_uri=CELL_ANNOTATION_SCHEMA.matrix_file_id, domain=None, range=Optional[Union[str, URIorCURIE]])
 
 slots.title = Slot(uri=CELL_ANNOTATION_SCHEMA.title, name="title", curie=CELL_ANNOTATION_SCHEMA.curie('title'),
                    model_uri=CELL_ANNOTATION_SCHEMA.title, domain=None, range=str)

src/cell_annotation_schema/datamodel/cell_annotation_schema.py

@@ -1,5 +1,5 @@
 # Auto generated from None by pythongen.py version: 0.0.1
-# Generation date: 2024-09-23T16:32:46
+# Generation date: 2024-10-03T17:13:02
 # Schema: General_Cell_Annotation_Open_Standard
 #
 # id: https://cellular-semantics.sanger.ac.uk/ontology/CAS
@@ -34,6 +34,7 @@
 # Namespaces
 CAS = CurieNamespace('CAS', 'https://cellular-semantics.sanger.ac.uk/ontology/CAS/')
 CL = CurieNamespace('CL', 'http://purl.obolibrary.org/obo/CL_')
+CELLXGENE_DATASET = CurieNamespace('CellXGene_dataset', 'https://cellxgene.cziscience.com/datasets/')
 IAO = CurieNamespace('IAO', 'http://purl.obolibrary.org/obo/IAO_')
 PCL = CurieNamespace('PCL', 'http://purl.obolibrary.org/obo/PCL_')
 RO = CurieNamespace('RO', 'http://purl.obolibrary.org/obo/RO_')
@@ -243,7 +244,7 @@ class Taxonomy(YAMLRoot):
     author_name: str = None
     labelsets: Union[Union[dict, Labelset], List[Union[dict, Labelset]]] = None
     annotations: Union[Union[dict, Annotation], List[Union[dict, Annotation]]] = None
-    matrix_file_id: Optional[str] = None
+    matrix_file_id: Optional[Union[str, URIorCURIE]] = None
     description: Optional[str] = None
     cellannotation_schema_version: Optional[str] = None
     cellannotation_timestamp: Optional[str] = None
@@ -276,8 +277,8 @@ def __post_init__(self, *_: List[str], **kwargs: Dict[str, Any]):
             self.annotations = [self.annotations] if self.annotations is not None else []
         self.annotations = [v if isinstance(v, Annotation) else Annotation(**as_dict(v)) for v in self.annotations]
 
-        if self.matrix_file_id is not None and not isinstance(self.matrix_file_id, str):
-            self.matrix_file_id = str(self.matrix_file_id)
+        if self.matrix_file_id is not None and not isinstance(self.matrix_file_id, URIorCURIE):
+            self.matrix_file_id = URIorCURIE(self.matrix_file_id)
 
         if self.description is not None and not isinstance(self.description, str):
             self.description = str(self.description)
@@ -14023,7 +14024,7 @@ class slots:
                    model_uri=CELL_ANNOTATION_SCHEMA.author_annotation_fields, domain=None, range=Optional[Union[dict, Any]])
 
 slots.matrix_file_id = Slot(uri=CELL_ANNOTATION_SCHEMA.matrix_file_id, name="matrix_file_id", curie=CELL_ANNOTATION_SCHEMA.curie('matrix_file_id'),
-                   model_uri=CELL_ANNOTATION_SCHEMA.matrix_file_id, domain=None, range=Optional[str])
+                   model_uri=CELL_ANNOTATION_SCHEMA.matrix_file_id, domain=None, range=Optional[Union[str, URIorCURIE]])
 
 slots.title = Slot(uri=CELL_ANNOTATION_SCHEMA.title, name="title", curie=CELL_ANNOTATION_SCHEMA.curie('title'),
                    model_uri=CELL_ANNOTATION_SCHEMA.title, domain=None, range=str)

src/cell_annotation_schema/generator/dataclassgen.py

@@ -100,10 +100,9 @@ def get_py_instance(instance_dict, schema_name, schema_def, root_class=None):
 
     for annotation in py_inst.annotations:
         # fix the author_annotation_fields in the json representation to be a string
-        if annotation.author_annotation_fields:
+        if annotation.author_annotation_fields and isinstance(annotation.author_annotation_fields, str):
             deserialised = json.loads(annotation.author_annotation_fields)
-            # annotation.author_annotation_fields = deserialised
-            annotation.author_annotation_fields = {"Cluster ID": "4"}
+            annotation.author_annotation_fields = deserialised
     return py_inst
 
 

src/cell_annotation_schema/ontology/data.py

@@ -7,11 +7,13 @@
 from cell_annotation_schema.ontology.schema import DEFAULT_PREFIXES, CAS_NAMESPACE
 from cell_annotation_schema.file_utils import get_json_from_file
 from cell_annotation_schema.generator.dataclassgen import get_py_instance, get_root_class
+from cell_annotation_schema.ontology.dumpers import rdflib_dumper
+from cell_annotation_schema.curie_utils import CurieToIriConverter
 
 from linkml_runtime.linkml_model import SchemaDefinition
 from linkml_runtime import SchemaView
 from linkml_runtime.loaders import yaml_loader
-from linkml_runtime.dumpers import rdflib_dumper
+# from linkml_runtime.dumpers import rdflib_dumper
 from linkml.validator import Validator
 
 CELL_RELATION = "has_cellid"
@@ -54,6 +56,7 @@ def dump_to_rdf(
     if validate:
         validate_data(schema_def, instance)
     instance = serialise_author_annotation(instance)
+    instance = resolve_matrix_file(instance)
 
     root_class = get_root_class(schema_name)
     py_inst = get_py_instance(instance, None, schema_def, root_class)
@@ -209,3 +212,17 @@ def serialise_author_annotation(instance: dict) -> dict:
         if "author_annotation_fields" in annotation and annotation["author_annotation_fields"]:
             annotation["author_annotation_fields"] = json.dumps(annotation["author_annotation_fields"])
     return instance
+
+
+def resolve_matrix_file(instance: dict) -> dict:
+    """
+    Resolves the matrix file curie to iri in the instance.
+    Args:
+        instance: The instance to be updated.
+    Returns:
+        The updated instance.
+    """
+    if "matrix_file_id" in instance and instance["matrix_file_id"]:
+        converter = CurieToIriConverter()
+        instance["matrix_file_id"] = converter.curie_to_iri(str(instance["matrix_file_id"])) + ".cxg/"
+    return instance

src/cell_annotation_schema/ontology/dumpers/__init__.py

@@ -0,0 +1,13 @@
+from linkml_runtime.dumpers.json_dumper import JSONDumper
+from linkml_runtime.dumpers.rdf_dumper import RDFDumper
+from cell_annotation_schema.ontology.dumpers.rdflib_dumper import RDFLibDumper
+from linkml_runtime.dumpers.tsv_dumper import TSVDumper
+from linkml_runtime.dumpers.yaml_dumper import YAMLDumper
+from linkml_runtime.dumpers.csv_dumper import CSVDumper
+
+json_dumper = JSONDumper()
+rdf_dumper = RDFDumper()
+rdflib_dumper = RDFLibDumper()
+yaml_dumper = YAMLDumper()
+csv_dumper = CSVDumper()
+tsv_dumper = TSVDumper()