[WIP] feat(ontologies): add ontology terms/resources models/types/instances

README.md

@@ -118,6 +118,12 @@ All Bento channels are prefixed with `bento.`.
 `logging` contains helper functions for standardized Bento logging configuration
 and formatting.
 
+### `ontologies`
+
+`ontologies` contains models, types, and helpers for working with ontology terms,
+especially in the context of terms which must be eventually ingested into 
+[Katsu](https://github.com/bento-platform/katsu).
+
 ### `responses`
 
 `responses` contains standardized error message-generating functions 

bento_lib/__init__.py

@@ -4,6 +4,7 @@
 from . import auth
 from . import drs
 from . import events
+from . import ontologies
 from . import schemas
 from . import search
 from . import service_info
@@ -12,5 +13,15 @@
 
 __version__ = metadata.version(__name__)
 __all__ = [
-    "__version__", "apps", "auth", "drs", "events", "schemas", "search", "service_info", "streaming", "workflows"
+    "__version__",
+    "apps",
+    "auth",
+    "drs",
+    "events",
+    "ontologies",
+    "schemas",
+    "search",
+    "service_info",
+    "streaming",
+    "workflows"
 ]

bento_lib/ontologies/common_resources.py

@@ -0,0 +1,120 @@
+from .models import OntologyResource, VersionedOntologyResource
+
+__all__ = [
+    # EFO
+    "EFO",
+    "EFO_3_69_0",
+    # MONDO
+    "MONDO",
+    "MONDO_2024_09_03",
+    # NCBITaxon
+    "NCBI_TAXON",
+    "NCBI_TAXON_2024_07_03",
+    # NCIT
+    "NCIT",
+    "NCIT_2024_05_07",
+    # OBI
+    "OBI",
+    "OBI_2024_06_10",
+    # SO
+    "SO",
+    # UBERON
+    "UBERON",
+]
+
+
+def _versioned(ont: OntologyResource, url: str, version: str) -> VersionedOntologyResource:
+    return VersionedOntologyResource(
+        **ont.model_dump(include={"id", "name", "namespace_prefix", "iri_prefix"}),
+        url=url,
+        version=version,
+    )
+
+
+# === EFO ==============================================================================================================
+
+EFO = OntologyResource(
+    id="efo",
+    name="Experimental Factor Ontology",
+    namespace_prefix="EFO",
+    iri_prefix="http://www.ebi.ac.uk/efo/EFO_",
+    url="http://www.ebi.ac.uk/efo/efo.owl",
+)
+EFO_3_69_0 = _versioned(EFO, "http://www.ebi.ac.uk/efo/releases/v3.69.0/efo.owl", version="3.69.0")
+
+# === MONDO ============================================================================================================
+
+MONDO = OntologyResource(
+    id="mondo",
+    name="Mondo Disease Ontology",
+    namespace_prefix="MONDO",
+    iri_prefix="http://purl.obolibrary.org/obo/MONDO_",
+    url="http://purl.obolibrary.org/obo/mondo.owl",
+)
+MONDO_2024_09_03 = _versioned(
+    MONDO,
+    url="http://purl.obolibrary.org/obo/mondo/releases/2024-09-03/mondo.owl",
+    version="2024-09-03",
+)
+
+# === NCBITaxon ========================================================================================================
+
+NCBI_TAXON = OntologyResource(
+    id="ncbitaxon",
+    name="NCBI organismal classification",
+    namespace_prefix="NCBITaxon",
+    iri_prefix="http://purl.obolibrary.org/obo/NCBITaxon_",
+    url="http://purl.obolibrary.org/obo/ncbitaxon.owl",
+)
+NCBI_TAXON_2024_07_03 = _versioned(
+    NCBI_TAXON,
+    url="http://purl.obolibrary.org/obo/ncbitaxon/2024-07-03/ncbitaxon.owl",
+    version="2024-07-03",
+)
+
+# === NCIT =============================================================================================================
+
+NCIT = OntologyResource(
+    id="ncit",
+    name="NCI Thesaurus OBO Edition",
+    namespace_prefix="NCIT",
+    iri_prefix="http://purl.obolibrary.org/obo/NCIT_",
+    url="http://purl.obolibrary.org/obo/ncit.owl",
+)
+NCIT_2024_05_07 = _versioned(
+    NCIT,
+    url="http://purl.obolibrary.org/obo/ncit/releases/2024-05-07/ncit.owl",
+    version="2024-05-07",
+)
+
+# === OBI ==============================================================================================================
+
+OBI = OntologyResource(
+    id="obi",
+    name="Ontology for Biomedical Investigations",
+    namespace_prefix="OBI",
+    iri_prefix="http://purl.obolibrary.org/obo/OBI_",
+    url="http://purl.obolibrary.org/obo/obi.owl",
+)
+OBI_2024_06_10 = _versioned(OBI, url="http://purl.obolibrary.org/obo/obi/2024-06-10/obi.owl", version="2024-06-10")
+
+# === SO ===============================================================================================================
+
+SO = OntologyResource(
+    id="so",
+    name="Sequence types and features ontology",
+    namespace_prefix="SO",
+    iri_prefix="http://purl.obolibrary.org/obo/SO_",
+    url="http://purl.obolibrary.org/obo/so.owl",
+)
+SO_2024_06_05 = _versioned(SO, url="http://purl.obolibrary.org/obo/so/2024-06-05/so.owl", version="2024-06-05")
+
+# === UBERON ===========================================================================================================
+
+UBERON = OntologyResource(
+    id="uberon",
+    name="Uberon multi-species anatomy ontology",
+    namespace_prefix="UBERON",
+    iri_prefix="http://purl.obolibrary.org/obo/UBERON_",
+    url="http://purl.obolibrary.org/obo/uberon.owl",
+)

bento_lib/ontologies/common_terms.py

@@ -0,0 +1,31 @@
+from .common_resources import NCBI_TAXON, OBI, SO
+
+__all__ = [
+    # NCBITaxon
+    "NCBI_TAXON_HOMO_SAPIENS",
+    "NCBI_TAXON_MUS_MUSCULUS",
+    # OBI
+    "OBI_16S_RRNA_ASSAY",
+    "OBI_RNA_SEQ_ASSAY",
+    "OBI_PROTEOMIC_PROFILING_BY_ARRAY_ASSAY",
+    "OBI_WHOLE_GENOME_SEQUENCING_ASSAY",
+    # SO
+    "SO_GENOMIC_DNA",
+]
+
+
+# === NCBITaxon ========================================================================================================
+
+NCBI_TAXON_HOMO_SAPIENS = NCBI_TAXON.make_term("NCBITaxon:9606", "Homo sapiens")
+NCBI_TAXON_MUS_MUSCULUS = NCBI_TAXON.make_term("NCBITaxon:10090", "Mus musculus")
+
+# === OBI ==============================================================================================================
+
+OBI_16S_RRNA_ASSAY = OBI.make_term("OBI:0002763", "16s ribosomal gene sequencing assay")
+OBI_RNA_SEQ_ASSAY = OBI.make_term("OBI:0001271", "RNA-seq assay")
+OBI_PROTEOMIC_PROFILING_BY_ARRAY_ASSAY = OBI.make_term("OBI:0001318", "proteomic profiling by array assay")
+OBI_WHOLE_GENOME_SEQUENCING_ASSAY = OBI.make_term("OBI:0002117", "whole genome sequencing assay")
+
+# === SO ===============================================================================================================
+
+SO_GENOMIC_DNA = SO.make_term("SO:0000991", "genomic DNA")

bento_lib/ontologies/models.py

@@ -0,0 +1,58 @@
+from pydantic import BaseModel, Field, HttpUrl
+from typing import Annotated
+
+from .types import PhenoV2Resource, PhenoV2OntologyClassDict
+
+NC_NAME_PATTERN = r"^[a-zA-Z_][a-zA-Z0-9.\-_]*$"
+CURIE_PATTERN = r"^[a-zA-Z_][a-zA-Z0-9.\-_]*:[a-zA-Z0-9.\-_]+$"
+
+
+class OntologyResource(BaseModel):
+    """
+    Inspired by the Phenopackets v2 Resource model:
+    https://phenopacket-schema.readthedocs.io/en/latest/resource.html
+    """
+
+    # From Phenopackets v2: "For OBO ontologies, the value of this string MUST always be the official OBO ID, which is
+    #   always equivalent to the ID prefix in lower case. Examples: hp, go, mp, mondo Consult http://obofoundry.org for
+    #   a complete list. For other resources which do not use native CURIE identifiers (e.g. SNOMED, UniProt, ClinVar),
+    #   use the prefix in identifiers.org."
+    id: str
+
+    # From Phenopackets v2: "The name of the ontology referred to by the id element, for example, The Human Phenotype
+    #   Ontology. For OBO Ontologies, the value of this string SHOULD be the same as the title field on
+    #   http://obofoundry.org. Other resources should use the official title for that resource. Note that this field is
+    #   purely for information purposes and software should not encode any assumptions."
+    name: str
+    url: HttpUrl
+    # From Phenopackets v2: "The prefix used in the CURIE of an OntologyClass e.g. HP, MP, ECO for example an HPO term
+    #   will have a CURIE like this - HP:0012828 which should be used in combination with the iri_prefix to form a
+    #   fully-resolvable IRI."
+    # Since we use it in a CURIE prefix context, it must match a valid NCName:
+    # https://www.w3.org/TR/1999/REC-xml-names-19990114/#NT-NCName
+    namespace_prefix: Annotated[str, Field(pattern=NC_NAME_PATTERN)]
+    iri_prefix: HttpUrl
+
+    def make_term(self, id_: str, label: str) -> "OntologyTerm":
+        return OntologyTerm(ontology=self, id=id_, label=label)
+
+
+class VersionedOntologyResource(OntologyResource):
+    version: str
+
+    def to_phenopackets_repr(self) -> PhenoV2Resource:
+        return self.model_dump(mode="json", include={"id", "version", "name", "url", "namespace_prefix", "iri_prefix"})
+
+
+class OntologyTerm(BaseModel):
+    """
+    Inspired by the Phenopackets v2 OntologyClass model:
+    https://phenopacket-schema.readthedocs.io/en/latest/ontologyclass.html
+    """
+
+    ontology: VersionedOntologyResource
+    id: Annotated[str, Field(pattern=CURIE_PATTERN)]
+    label: str
+
+    def to_phenopackets_repr(self) -> PhenoV2OntologyClassDict:
+        return self.model_dump(mode="json", include={"id", "label"})

bento_lib/ontologies/types.py

@@ -0,0 +1,20 @@
+from typing import TypedDict
+
+__all__ = [
+    "PhenoV2Resource",
+    "PhenoV2OntologyClassDict",
+]
+
+
+class PhenoV2Resource(TypedDict):
+    id: str
+    name: str
+    url: str
+    version: str
+    namespace_prefix: str
+    iri_prefix: str
+
+
+class PhenoV2OntologyClassDict(TypedDict):
+    id: str
+    label: str