Merge pull request #58 from databio/dev

Release 0.7.0

bbconf/_version.py

@@ -1 +1 @@
-__version__ = "0.6.1"
+__version__ = "0.7.0"

bbconf/bbagent.py

@@ -42,7 +42,6 @@ def bedset(self) -> BedAgentBedSet:
     def objects(self) -> BBObjects:
         return self._objects
 
-    @cached_property
     def get_stats(self) -> StatsReturn:
         """
         Get statistics for a bed file

bbconf/config_parser/bedbaseconfig.py

@@ -10,7 +10,7 @@
 import yacman
 import zarr
 from botocore.exceptions import BotoCoreError, EndpointConnectionError
-from geniml.region2vec import Region2VecExModel
+from geniml.region2vec.main import Region2VecExModel
 from geniml.search import BED2BEDSearchInterface, QdrantBackend, Text2BEDSearchInterface
 from geniml.search.query2vec import BED2Vec, Text2Vec
 from pephubclient import PEPHubClient

bbconf/db_utils.py

@@ -3,7 +3,16 @@
 from typing import List, Optional
 
 import pandas as pd
-from sqlalchemy import TIMESTAMP, BigInteger, ForeignKey, Result, Select, event, select
+from sqlalchemy import (
+    TIMESTAMP,
+    BigInteger,
+    ForeignKey,
+    Result,
+    Select,
+    event,
+    select,
+    UniqueConstraint,
+)
 from sqlalchemy.dialects.postgresql import JSON
 from sqlalchemy.engine import URL, Engine, create_engine
 from sqlalchemy.event import listens_for
@@ -99,6 +108,10 @@ class Bed(Base):
         "BedFileBedSetRelation", back_populates="bedfile", cascade="all, delete-orphan"
     )
 
+    annotations: Mapped["BedMetadata"] = relationship(
+        back_populates="bed", cascade="all, delete-orphan", lazy="joined"
+    )
+
     stats: Mapped["BedStats"] = relationship(
         back_populates="bed", cascade="all, delete-orphan"
     )
@@ -114,6 +127,74 @@ class Bed(Base):
     )
     license_mapping: Mapped["License"] = relationship("License", back_populates="bed")
 
+    ref_classifier: Mapped["GenomeRefStats"] = relationship(
+        "GenomeRefStats", back_populates="bed", cascade="all, delete-orphan"
+    )
+
+
+class BedMetadata(Base):
+    __tablename__ = "bed_metadata"
+
+    species_name: Mapped[str] = mapped_column(default=None, comment="Organism name")
+    species_id: Mapped[str] = mapped_column(
+        default=None, nullable=True, comment="Organism taxon id"
+    )
+
+    genotype: Mapped[str] = mapped_column(
+        default=None, nullable=True, comment="Genotype of the sample"
+    )
+    phenotype: Mapped[str] = mapped_column(
+        default=None, nullable=True, comment="Phenotype of the sample"
+    )
+
+    cell_type: Mapped[str] = mapped_column(
+        default=None,
+        nullable=True,
+        comment="Specific kind of cell with distinct characteristics found in an organism. e.g. Neurons, Hepatocytes, Adipocytes",
+    )
+    cell_line: Mapped[str] = mapped_column(
+        default=None,
+        nullable=True,
+        comment="Population of cells derived from a single cell and cultured in the lab for extended use, e.g. HeLa, HepG2, k562",
+    )
+    tissue: Mapped[str] = mapped_column(
+        default=None, nullable=True, comment="Tissue type"
+    )
+    library_source: Mapped[str] = mapped_column(
+        default=None,
+        nullable=True,
+        comment="Library source (e.g. genomic, transcriptomic)",
+    )
+    assay: Mapped[str] = mapped_column(
+        default=None, nullable=True, comment="Experimental protocol (e.g. ChIP-seq)"
+    )
+    antibody: Mapped[str] = mapped_column(
+        default=None, nullable=True, comment="Antibody used in the assay"
+    )
+    target: Mapped[str] = mapped_column(
+        default=None, nullable=True, comment="Target of the assay (e.g. H3K4me3)"
+    )
+    treatment: Mapped[str] = mapped_column(
+        default=None,
+        nullable=True,
+        comment="Treatment of the sample (e.g. drug treatment)",
+    )
+
+    global_sample_id: Mapped[str] = mapped_column(
+        default=None, nullable=True, comment="Global sample identifier. e.g. GSM000"
+    )
+    global_experiment_id: Mapped[str] = mapped_column(
+        default=None, nullable=True, comment="Global experiment identifier. e.g. GSE000"
+    )
+
+    id: Mapped[str] = mapped_column(
+        ForeignKey("bed.id", ondelete="CASCADE"),
+        primary_key=True,
+        index=True,
+    )
+
+    bed: Mapped["Bed"] = relationship("Bed", back_populates="annotations")
+
 
 class BedStats(Base):
     __tablename__ = "bed_stats"
@@ -283,6 +364,32 @@ class License(Base):
     bed: Mapped[List["Bed"]] = relationship("Bed", back_populates="license_mapping")
 
 
+class GenomeRefStats(Base):
+    __tablename__ = "genome_ref_stats"
+
+    id: Mapped[int] = mapped_column(primary_key=True, index=True)
+
+    bed_id: Mapped[str] = mapped_column(
+        ForeignKey("bed.id", ondelete="CASCADE"),
+        index=True,
+        nullable=False,
+    )
+    provided_genome: Mapped[str]
+    compared_genome: Mapped[str] = mapped_column(
+        nullable=False, comment="Compared Genome"
+    )
+
+    xs: Mapped[float] = mapped_column(nullable=True, default=None)
+    oobr: Mapped[float] = mapped_column(nullable=True, default=None)
+    sequence_fit: Mapped[float] = mapped_column(nullable=True, default=None)
+    assigned_points: Mapped[int] = mapped_column(nullable=False)
+    tier_ranking: Mapped[int] = mapped_column(nullable=False)
+
+    bed: Mapped["Bed"] = relationship("Bed", back_populates="ref_classifier")
+
+    __table_args__ = (UniqueConstraint("bed_id", "compared_genome"),)
+
+
 @listens_for(Universes, "after_insert")
 @listens_for(Universes, "after_update")
 def add_bed_universe(mapper, connection, target):

bbconf/models/bed_models.py

@@ -98,6 +98,54 @@ class BedPEPHub(BaseModel):
     model_config = ConfigDict(extra="allow", populate_by_name=True)
 
 
+class StandardMeta(BaseModel):
+    """
+    Standardized Bed file metadata
+    """
+
+    species_name: str = Field(
+        default="", description="Name of species. e.g. Homo sapiens.", alias="organism"
+    )
+    species_id: str = ""
+    genotype: str = Field("", description="Genotype of the sample")
+    phenotype: str = Field("", description="Phenotype of the sample")
+    description: Union[str, None] = ""
+
+    cell_type: str = Field(
+        "",
+        description="specific kind of cell with distinct characteristics found in an organism. e.g. Neurons, Hepatocytes, Adipocytes",
+    )
+    cell_line: str = Field(
+        "",
+        description="population of cells derived from a single cell and cultured in the lab for extended use, e.g. HeLa, HepG2, k562",
+    )
+    tissue: str = Field("", description="Tissue type")
+
+    library_source: str = Field(
+        "", description="Library source (e.g. genomic, transcriptomic)"
+    )
+    assay: str = Field(
+        "", description="Experimental protocol (e.g. ChIP-seq)", alias="exp_protocol"
+    )
+    antibody: str = Field("", description="Antibody used in the assay")
+    target: str = Field("", description="Target of the assay (e.g. H3K4me3)")
+    treatment: str = Field(
+        "", description="Treatment of the sample (e.g. drug treatment)"
+    )
+
+    global_sample_id: str = Field(
+        "", description="Global sample identifier. e.g. GSM000"
+    )  # excluded in training
+    global_experiment_id: str = Field(
+        "", description="Global experiment identifier. e.g. GSE000"
+    )  # excluded in training
+
+    model_config = ConfigDict(
+        populate_by_name=True,
+        extra="ignore",
+    )
+
+
 class BedPEPHubRestrict(BedPEPHub):
 
     model_config = ConfigDict(extra="ignore")
@@ -111,6 +159,7 @@ class BedMetadataBasic(BedClassification):
     last_update_date: Optional[datetime.datetime] = None
     is_universe: Optional[bool] = False
     license_id: Optional[str] = DEFAULT_LICENSE
+    annotation: Optional[StandardMeta] = None
 
 
 class UniverseMetadata(BaseModel):
@@ -124,7 +173,7 @@ class BedSetMinimal(BaseModel):
     description: Union[str, None] = None
 
 
-class BedMetadata(BedMetadataBasic):
+class BedMetadataAll(BedMetadataBasic):
     stats: Union[BedStatsModel, None] = None
     plots: Union[BedPlots, None] = None
     files: Union[BedFiles, None] = None
@@ -171,3 +220,15 @@ class TokenizedPathResponse(BaseModel):
     universe_id: str
     file_path: str
     endpoint_url: str
+
+
+class RefGenValidModel(BaseModel):
+    provided_genome: str
+    compared_genome: str
+    xs: float = 0.0
+    oobr: Union[float, None] = None
+    sequence_fit: Union[float, None] = None
+    assigned_points: int
+    tier_ranking: int
+
+    model_config = ConfigDict(extra="forbid")