Create single result row even if multiple clinvar entries (#565)

HISTORY.rst

@@ -13,6 +13,7 @@ End-User Summary
 - Extra annotations in export completed and tested (#495).
 - Fixed bug where Exac and thousand genomes settings were not shown in frequency tab for GRCh37 (#597).
 - Form template reports error if genomebuild variable is not set (#607).
+- Create single result row even if multiple clinvar entries (#565).
 
 Full Change List
 ================
@@ -21,6 +22,7 @@ Full Change List
 - Fixing issue with sync-from-remote when no remote is defined (#570).
 - Fixed bug where Exac and thousand genomes settings were not shown in frequency tab for GRCh37 (#597).
 - Form template reports error if genomebuild variable is not set (#607).
+- Create single result row even if multiple clinvar entries (#565).
 
 -----------------
 v1.2.1 (anthenea)

variants/file_export.py

@@ -110,7 +110,7 @@ def to_str(val):
     ("gnomad_oe_lof", "Gnomad constrains lof observed/expected", float),
     ("gnomad_oe_lof_upper", "Gnomad constrains lof observed/expected upper", float),
     ("gnomad_oe_lof_lower", "Gnomad constrains lof observed/expected lower", float),
-    ("pathogenicity_summary", "ClinVar pathogenicity summary", str),
+    ("pathogenicity_summary_arr", "ClinVar pathogenicity summary", str),
 )
 if settings.KIOSK_MODE:
     HEADER_FIXED = tuple(filter(lambda x: not x[0].startswith("inhouse_"), HEADER_FIXED))
@@ -123,8 +123,8 @@ def to_str(val):
 
 #: Names of the pathogenicity scoring header columns.
 HEADERS_PATHO_SCORES = (
-    ("pathogenicity_score", "Pathogenicity Score", float),
-    ("pathogenicity_rank", "Pathogenicity Rank", int),
+    ("pathogenicity_score_arr", "Pathogenicity Score", float),
+    ("pathogenicity_rank_arr", "Pathogenicity Rank", int),
 )
 
 HEADERS_TRANSCRIPTS = (("transcripts", "Transcript ids", str),)

variants/queries.py

@@ -9,8 +9,9 @@
 from django.conf import settings
 from django.core.exceptions import ImproperlyConfigured
 from sqlalchemy import Table, true, column, union, literal_column, delete, tuple_
+from sqlalchemy.dialects.postgresql import ARRAY
 from sqlalchemy.sql import select, func, and_, not_, or_, cast
-from sqlalchemy.types import ARRAY, VARCHAR, Integer, Float
+from sqlalchemy.types import VARCHAR, Integer, Float, String
 import sqlparse
 
 from clinvar.models import Clinvar
@@ -360,16 +361,23 @@ def extend_selectable(self, query_parts):
 class ExtendQueryPartsClinvarJoin(ExtendQueryPartsBase):
     def __init__(self, *args, **kwargs):
         super().__init__(*args, **kwargs)
+        self._col_names = (
+            "variation_type",
+            "vcv",
+            "point_rating",
+            "pathogenicity",
+            "review_status",
+            "pathogenicity_summary",
+            "details",
+        )
+
         self.subquery = (
             select(
-                (
-                    Clinvar.sa.variation_type,
-                    Clinvar.sa.vcv,
-                    Clinvar.sa.point_rating,
-                    Clinvar.sa.pathogenicity,
-                    Clinvar.sa.review_status,
-                    Clinvar.sa.pathogenicity_summary,
-                    Clinvar.sa.details,
+                tuple(
+                    func.array_agg(
+                        getattr(Clinvar.sa, name), type_=ARRAY(String(length=128))
+                    ).label(f"{name}_arr")
+                    for name in self._col_names
                 )
             )
             .select_from(Clinvar.sa)
@@ -382,12 +390,8 @@ def extend_selectable(self, query_parts):
 
     def extend_fields(self, _query_parts):
         return [
-            self.subquery.c.variation_type,
-            self.subquery.c.vcv,
-            self.subquery.c.point_rating,
-            self.subquery.c.pathogenicity,
-            self.subquery.c.pathogenicity_summary,
-            self.subquery.c.details,
+            func.coalesce(getattr(self.subquery.c, f"{name}_arr"), []).label(f"{name}_arr")
+            for name in self._col_names
         ]
 
 
@@ -417,7 +421,10 @@ def _build_significance_term(self):
             return True
         for patho_key in self.patho_keys:
             if self.kwargs.get("clinvar_include_%s" % patho_key):
-                terms.append(self.subquery.c.pathogenicity == patho_key.replace("_", " "))
+                # import pdb; pdb.set_trace()
+                terms.append(
+                    self.subquery.c.pathogenicity_arr.contains([patho_key.replace("_", " ")])
+                )
         return or_(*terms)
 
 

variants/tests/test_queries.py

@@ -4476,6 +4476,33 @@ def test_render_query_require_membership_include_benign(self):
         )
         self.assertEqual(res[0].start, self.small_vars[6].start)
 
+    def test_render_query_single_output_line_even_with_multiple_clinvar_annos(self):
+        # Add second ClinVar annotation
+        ClinvarFactory(
+            release=self.small_vars[1].release,
+            chromosome=self.small_vars[1].chromosome,
+            start=self.small_vars[1].start,
+            end=self.small_vars[1].end,
+            bin=self.small_vars[1].bin,
+            reference=self.small_vars[1].reference,
+            alternative=self.small_vars[1].alternative,
+            pathogenicity="pathogenic",
+        )
+        res = self.run_query(
+            self.query_class,
+            {
+                "genomic_region": [
+                    (
+                        self.small_vars[1].chromosome,
+                        self.small_vars[1].start - 1,
+                        self.small_vars[1].end + 1,
+                    )
+                ]
+            },
+            1,
+        )
+        self.assertEqual(res[0].start, self.small_vars[1].start)
+
 
 class RenderQueryTestCaseThreeClinvarFilter(CaseThreeClinvarFilterTestMixin, SupportQueryTestBase):
     """Test clinvar membership using RenderFilterQuery."""

variants/tests/test_views.py

@@ -1098,7 +1098,7 @@ def test_clinvar(self):
                 ),
                 {"filter_job_uuid": self.bgjob.sodar_uuid},
             )
-            self.assertEqual(response.context["result_rows"][1].pathogenicity, "pathogenic")
+            self.assertEqual(response.context["result_rows"][1].pathogenicity_arr, ["pathogenic"])
 
     def test_training_mode(self):
         with self.login(self.superuser):
@@ -2264,7 +2264,7 @@ def test_clinvar(self):
                 {"filter_job_uuid": self.bgjob.sodar_uuid},
             )
             self.assertEqual(response.status_code, 200)
-            self.assertEqual(response.context["result_rows"][4].pathogenicity, "pathogenic")
+            self.assertEqual(response.context["result_rows"][4].pathogenicity_arr, ["pathogenic"])
 
     @patch("django.conf.settings.VARFISH_ENABLE_CADD", True)
     @patch("django.conf.settings.VARFISH_CADD_REST_API_URL", "https://cadd.com")