Merge pull request #422 from Clinical-Genomics/n_intervals_by_build

Intervals count by build endpoint

CHANGELOG.md

@@ -1,4 +1,6 @@
 ## [unreleased]
+### Added
+- `/intervals/intervals_count_by_build` endpoint, which retuns the number of genes, transcripts, exons for each genome build
 ### Changed
 - Updated schug library to v1.10
 - Disable SQLAlchemy logger

src/chanjo2/crud/intervals.py

@@ -1,18 +1,12 @@
 import logging
-from typing import List, Optional, Union
+from typing import Dict, List, Optional, Union
 
-from sqlalchemy import delete, or_, text
+from sqlalchemy import delete, func, or_, text
 from sqlalchemy.orm import Session, query
 from sqlalchemy.sql.expression import Delete
 
 from chanjo2.models import SQLExon, SQLGene, SQLTranscript
-from chanjo2.models.pydantic_models import (
-    Builds,
-    ExonBase,
-    GeneBase,
-    TranscriptBase,
-    TranscriptTag,
-)
+from chanjo2.models.pydantic_models import Builds, TranscriptBase, TranscriptTag
 
 LOG = logging.getLogger(__name__)
 
@@ -192,3 +186,20 @@ def bulk_insert_exons(db: Session, exons: List[SQLExon]) -> None:
     """Bulk insert exons into the database."""
     db.bulk_save_objects(exons)
     db.commit()
+
+
+def get_interval_counts(db: Session) -> Dict:
+    counts = {}
+    for build in Builds.get_enum_values():
+        counts[build] = {
+            "number_of_genes": db.query(func.count(SQLGene.id))
+            .filter(SQLGene.build == build)
+            .scalar(),
+            "number_of_transcripts": db.query(func.count(SQLTranscript.id))
+            .filter(SQLTranscript.build == build)
+            .scalar(),
+            "number_of_exons": db.query(func.count(SQLExon.id))
+            .filter(SQLExon.build == build)
+            .scalar(),
+        }
+    return counts

src/chanjo2/endpoints/intervals.py

@@ -1,19 +1,19 @@
-from typing import Iterator, List, Optional
+from typing import Iterator, List
 
 from fastapi import APIRouter, Depends, HTTPException, Response, status
 from fastapi.responses import JSONResponse
 from sqlalchemy.orm import Session
 
 from chanjo2.constants import MULTIPLE_PARAMS_NOT_SUPPORTED_MSG
-from chanjo2.crud.intervals import get_gene_intervals, get_genes
+from chanjo2.crud.intervals import get_gene_intervals, get_genes, get_interval_counts
 from chanjo2.dbutil import get_session
 from chanjo2.meta.handle_bed import resource_lines
 from chanjo2.meta.handle_load_intervals import (
     update_exons,
     update_genes,
     update_transcripts,
 )
-from chanjo2.models import SQLExon, SQLGene, SQLTranscript
+from chanjo2.models import SQLExon, SQLTranscript
 from chanjo2.models.pydantic_models import (
     Builds,
     ExonBase,
@@ -184,3 +184,9 @@ async def exons(query: GeneIntervalQuery, session: Session = Depends(get_session
         limit=query.limit if nr_filters == 0 else None,
         interval_type=SQLExon,
     )
+
+
+@router.get("/intervals/intervals_count_by_build", response_model=dict)
+def intervals_count_by_build(session: Session = Depends(get_session)):
+    """Returns the number of genes, transcripts and exons available in the database for each genome build."""
+    return get_interval_counts(db=session)

src/chanjo2/models/pydantic_models.py

@@ -37,6 +37,11 @@ class IntervalType(str, Enum):
     EXONS = "exons"
     CUSTOM = "custom_intervals"
 
+    @staticmethod
+    def get_enum_values() -> List[str]:
+        """Returns the values of the available interval types."""
+        return [member.value for member in IntervalType]
+
 
 class TranscriptTag(str, Enum):
     REFSEQ_MRNA = "refseq_mrna"

tests/conftest.py

@@ -62,6 +62,7 @@ class Endpoints(str):
     TRANSCRIPTS = "/intervals/transcripts"
     LOAD_EXONS = "/intervals/load/exons/"
     EXONS = "/intervals/exons"
+    INTERVALS_BY_BUILD = "/intervals/intervals_count_by_build"
     INTERVAL_COVERAGE = "/coverage/d4/interval/"
     INTERVALS_FILE_COVERAGE = "/coverage/d4/interval_file/"
     GENES_COVERAGE_SUMMARY = "/coverage/d4/genes/summary"

tests/src/chanjo2/endpoints/test_intervals.py

@@ -6,7 +6,13 @@
 
 from chanjo2.constants import MULTIPLE_PARAMS_NOT_SUPPORTED_MSG
 from chanjo2.meta.handle_bed import resource_lines
-from chanjo2.models.pydantic_models import Builds, ExonBase, GeneBase, TranscriptBase
+from chanjo2.models.pydantic_models import (
+    Builds,
+    ExonBase,
+    GeneBase,
+    IntervalType,
+    TranscriptBase,
+)
 from chanjo2.populate_demo import (
     BUILD_EXONS_RESOURCE,
     BUILD_GENES_RESOURCE,
@@ -458,3 +464,23 @@ def test_exons_by_hgnc_symbols(
     assert response.status_code == status.HTTP_200_OK
     exons = response.json()
     assert ExonBase(**exons[0])
+
+
+def test_intervals_count_by_build(demo_client: TestClient, endpoints: Type):
+    """Tests the endpoint that returns the number of genes, transcripts and exons available in the database for each genome build."""
+
+    # GIVEN a populated demo database
+    # WHEN sending a GET request to the "intervals_count_by_build" endpoint
+    response: Response = demo_client.get(endpoints.INTERVALS_BY_BUILD)
+
+    # THEN response should be successful
+    assert response.status_code == status.HTTP_200_OK
+    result: dict = response.json()
+
+    # AND contain number of intervals by build
+    for build in Builds.get_enum_values():
+        for itype in IntervalType.get_enum_values():
+            if itype == "custom_intervals":
+                continue
+            count = result[build][f"number_of_{itype}"]
+            assert isinstance(count, int)