added cell line to subjects

.github/workflows/publish_dev.yml

@@ -33,5 +33,5 @@ jobs:
       - name: Create and upload schemas
         run: |
           uv sync --locked
-          uv run python -m biodata_schema.utils.json_writer --output $TEMP_DIR --attach-version
+          uv run python -m biodata_schema.utils.json_writer --output $TEMP_DIR --attach-version --draft
           uvx --from awscli aws s3 sync $TEMP_DIR s3://${AWS_DATA_SCHEMA_BUCKET}/$S3_PREFIX

.github/workflows/tag_and_publish_main.yml

@@ -66,5 +66,5 @@ jobs:
       - name: Create and upload schemas
         run: |
           uv sync --locked
-          uv run python -m biodata_schema.utils.json_writer --output $TEMP_DIR --attach-version
+          uv run python -m biodata_schema.utils.json_writer --output $TEMP_DIR --attach-version --draft
           uvx --from awscli aws s3 sync $TEMP_DIR s3://${AWS_DATA_SCHEMA_BUCKET}/$S3_PREFIX

pyproject.toml

@@ -14,7 +14,7 @@ readme = "README.md"
 dynamic = ["version"]
 
 dependencies = [
-    'biodata-models>=1.0.1,<2',
+    'biodata-models>=2.0.1,<3',
     'pydantic>=2.7, <3',
     'pydantic-extra-types',
     'tzdata',
@@ -25,11 +25,12 @@ dev = [
     'argparse',
     'codespell',
     'dictdiffer',
+    'jsonschema>=4.23,<5',
     'pydantic>=2.7, !=2.9.0, !=2.9.1',
     'pytest>=8.3',
     'pytest-cov>=6',
     'ruff>=0.11',
-    'semver',
+    'semver>=3.0.0',
 ]
 
 docs = [

src/biodata_schema/__init__.py

@@ -1,3 +1,3 @@
 """base module for biodata-schema"""
 
-__version__ = "3.0.0"
+__version__ = "4.0.0"

src/biodata_schema/base.py

@@ -30,6 +30,10 @@
 MAX_FILE_SIZE = 500 * 1024  # 500KB
 
 
+class DraftRequirement:
+    """Marker for fields required by the draft metadata schema."""
+
+
 def _coerce_naive_datetime(v: Any, handler: ValidatorFunctionWrapHandler) -> AwareDatetime:
     """Validator to wrap around AwareDatetime to set a default timezone as user's locale"""
     try:

src/biodata_schema/components/subjects.py

@@ -5,6 +5,7 @@
 from enum import Enum
 from typing import Annotated, List, Optional
 
+from biodata_models.cell_line import CellLineModel
 from biodata_models.organizations import Organization
 from biodata_models.pid_names import PIDName
 from biodata_models.species import Species, Strain
@@ -195,3 +196,16 @@ class CalibrationObject(DataModel):
     objects: Optional[list[Device]] = Field(
         default=None, title="Objects", description="For calibration objects that are built up from one or more devices."
     )
+
+
+class CellLine(DataModel):
+    """Description of a cultured cell line"""
+
+    cell_line_name: str = Field(..., title="Cell line name")
+    cell_line_type: CellLineModel = Field(..., title="Cell line type")
+    species: Species.ONE_OF = Field(..., title="Species")
+    protein: PIDName = Field(..., title="Protein labeled", description="Protein uses UniProt registry")
+    gene: PIDName = Field(..., title="Gene targeted", description="Gene uses NCBI taxonomy")
+    cell_structure: str = Field(..., title="Cell structure protein found in")  # TODO: ontology or enum in model?
+    fluorescent_protein: PIDName = Field(..., title="Fluorescent protein", description="Uses FPbase")
+    clone_number: Optional[int] = Field(default=None, title="Clone number")

src/biodata_schema/core/acquisition.py

@@ -18,6 +18,7 @@
     DataCoreModel,
     DataModel,
     DiscriminatedList,
+    DraftRequirement,
     GenericModel,
 )
 from biodata_schema.components.configs import (
@@ -389,7 +390,7 @@ class Acquisition(ProtocolListMixin, DataCoreModel):
     )
 
     # Acquisition metadata
-    acquisition_start_time: AwareDatetimeWithDefault = Field(
+    acquisition_start_time: Annotated[AwareDatetimeWithDefault, DraftRequirement] = Field(
         ...,
         title="Acquisition start time",
         description="During validation, timezone information will be moved into the acquisition_start_tz field.",

src/biodata_schema/core/data_description.py

@@ -1,7 +1,7 @@
 """Generic metadata classes for data"""
 
 import re
-from typing import List, Literal, Optional
+from typing import Annotated, List, Literal, Optional
 
 from biodata_models.data_name_patterns import (
     DataLevel,
@@ -15,7 +15,7 @@
 from biodata_models.organizations import Organization
 from pydantic import Field, SkipValidation, model_validator
 
-from biodata_schema.base import AwareDatetimeWithDefault, DataCoreModel, DataModel
+from biodata_schema.base import AwareDatetimeWithDefault, DataCoreModel, DataModel, DraftRequirement
 from biodata_schema.components.identifiers import Person
 
 
@@ -35,7 +35,7 @@ class DataDescription(DataCoreModel):
     _DESCRIBED_BY_URL = DataCoreModel._DESCRIBED_BY_BASE_URL.default + "biodata_schema/core/data_description.py"
     describedBy: str = Field(default=_DESCRIBED_BY_URL, json_schema_extra={"const": _DESCRIBED_BY_URL})
     schema_version: SkipValidation[Literal["3.0.0"]] = Field(default="3.0.0")
-    license: License = Field(default=License.CC_BY_40, title="License")
+    license: Annotated[License, DraftRequirement] = Field(default=License.CC_BY_40, title="License")
 
     subject_id: Optional[str] = Field(
         default=None,
@@ -89,7 +89,7 @@ class DataDescription(DataCoreModel):
         title="Investigators",
         min_length=1,
     )
-    project_name: str = Field(
+    project_name: Annotated[str, DraftRequirement] = Field(
         ...,
         pattern=DataRegex.NO_SPECIAL_CHARS_EXCEPT_SPACE.value,
         description="A name for a set of coordinated activities intended to achieve one or more objectives.",

src/biodata_schema/core/instrument.py

@@ -2,12 +2,12 @@
 
 import logging
 from datetime import date
-from typing import List, Literal, Optional
+from typing import Annotated, List, Literal, Optional
 
 from biodata_models.modalities import Modality
 from pydantic import Field, SkipValidation, field_validator, model_validator
 
-from biodata_schema.base import DataCoreModel, DiscriminatedList
+from biodata_schema.base import DataCoreModel, DiscriminatedList, DraftRequirement
 from biodata_schema.components.connections import Connection
 from biodata_schema.components.coordinates import CoordinateSystem
 from biodata_schema.components.devices import (
@@ -82,7 +82,7 @@ class Instrument(DataCoreModel):
 
     # instrument definition
     location: Optional[str] = Field(default=None, title="Location", description="Location of the instrument")
-    instrument_id: str = Field(
+    instrument_id: Annotated[str, DraftRequirement] = Field(
         ...,
         description="Unique instrument identifier",
         title="Instrument ID",

src/biodata_schema/core/metadata.py

@@ -5,7 +5,7 @@
 import logging
 import warnings
 from pathlib import Path
-from typing import Dict, List, Literal, Optional, Union, get_args
+from typing import Annotated, Dict, List, Literal, Optional, Union, get_args
 
 from biodata_models.modalities import Modality
 from pydantic import (
@@ -19,7 +19,7 @@
     model_validator,
 )
 
-from biodata_schema.base import DataCoreModel
+from biodata_schema.base import DataCoreModel, DraftRequirement
 from biodata_schema.components.identifiers import DatabaseIdentifiers
 from biodata_schema.components.subject_procedures import TrainingProtocol
 from biodata_schema.components.subjects import CalibrationObject
@@ -80,7 +80,7 @@ class Metadata(DataCoreModel):
         description="Name of the data asset.",
         title="Data Asset Name",
     )
-    location: str = Field(
+    location: Annotated[str, DraftRequirement] = Field(
         ...,
         title="Location",
         description="Current location of the data asset.",
@@ -296,6 +296,31 @@ def validate_calibration_object_tags(self):
 
         return self
 
+    @model_validator(mode="after")
+    def validate_subject_id_consistency(self):
+        """Validator to ensure procedures, acquisition, and data_description subject_id match subject.subject_id"""
+
+        if not self.subject:
+            return self
+
+        expected = self.subject.subject_id
+        mismatches = []
+        # getattr guards against objects built with model_construct() that omit subject_id
+        procedures_subject_id = getattr(self.procedures, "subject_id", None)
+        if procedures_subject_id is not None and procedures_subject_id != expected:
+            mismatches.append(f"procedures.subject_id={procedures_subject_id}")
+        acquisition_subject_id = getattr(self.acquisition, "subject_id", None)
+        if acquisition_subject_id is not None and acquisition_subject_id != expected:
+            mismatches.append(f"acquisition.subject_id={acquisition_subject_id}")
+        data_description_subject_id = getattr(self.data_description, "subject_id", None)
+        if data_description_subject_id is not None and data_description_subject_id != expected:
+            mismatches.append(f"data_description.subject_id={data_description_subject_id}")
+
+        if mismatches:
+            raise ValueError(f"subject_id mismatch with subject.subject_id={expected}: {', '.join(mismatches)}")
+
+        return self
+
     @model_validator(mode="after")
     def validate_training_protocol_references(self):
         """Validate that training_protocol_name in StimulusEpoch matches a TrainingProtocol in procedures"""

src/biodata_schema/core/subject.py

@@ -1,11 +1,17 @@
-"""schema for mostly mouse metadata"""
+"""schema for subject metadata"""
 
-from typing import Literal, Optional
+from typing import Annotated, Literal, Optional
 
 from pydantic import Field, SkipValidation
 
-from biodata_schema.base import DataCoreModel, Discriminated
-from biodata_schema.components.subjects import CalibrationObject, HumanSubject, MouseSubject, NonHumanPrimateSubject
+from biodata_schema.base import DataCoreModel, Discriminated, DraftRequirement
+from biodata_schema.components.subjects import (
+    CalibrationObject,
+    CellLine,
+    HumanSubject,
+    MouseSubject,
+    NonHumanPrimateSubject,
+)
 
 
 class Subject(DataCoreModel):
@@ -14,14 +20,14 @@ class Subject(DataCoreModel):
     _DESCRIBED_BY_URL = DataCoreModel._DESCRIBED_BY_BASE_URL.default + "biodata_schema/core/subject.py"
     describedBy: str = Field(default=_DESCRIBED_BY_URL, json_schema_extra={"const": _DESCRIBED_BY_URL})
     schema_version: SkipValidation[Literal["3.0.2"]] = Field(default="3.0.2")
-    subject_id: str = Field(
+    subject_id: Annotated[str, DraftRequirement] = Field(
         ...,
         description="Unique identifier for the subject of data acquisition",
         title="Subject ID",
     )
 
-    subject_details: Discriminated[MouseSubject | HumanSubject | NonHumanPrimateSubject | CalibrationObject] = Field(
-        ..., title="Subject Details"
-    )
+    subject_details: Discriminated[
+        MouseSubject | HumanSubject | NonHumanPrimateSubject | CellLine | CalibrationObject
+    ] = Field(..., title="Subject Details")
 
     notes: Optional[str] = Field(default=None, title="Notes")