Change the VariantType model - move publications to separate table

gene2phenotype_project/gene2phenotype_app/fixtures/lgd_variant_type.json

@@ -8,7 +8,6 @@
             "inherited": false,
             "de_novo": false,
             "unknown_inheritance": false,
-            "publication": 2,
             "is_deleted": 0
         }
     },
@@ -21,7 +20,6 @@
             "inherited": false,
             "de_novo": false,
             "unknown_inheritance": true,
-            "publication": 2,
             "is_deleted": 0
         }
     },
@@ -34,7 +32,6 @@
             "inherited": false,
             "de_novo": false,
             "unknown_inheritance": false,
-            "publication": 1,
             "is_deleted": 0
         }
     },
@@ -47,8 +44,31 @@
             "inherited": false,
             "de_novo": false,
             "unknown_inheritance": false,
-            "publication": 2,
             "is_deleted": 1
         }
+    },
+    {
+        "model": "gene2phenotype_app.lgdvarianttype",
+        "pk": 5,
+        "fields": {
+            "lgd": 11,
+            "variant_type_ot": 13,
+            "inherited": false,
+            "de_novo": false,
+            "unknown_inheritance": false,
+            "is_deleted": 0
+        }
+    },
+    {
+        "model": "gene2phenotype_app.lgdvarianttype",
+        "pk": 6,
+        "fields": {
+            "lgd": 10,
+            "variant_type_ot": 13,
+            "inherited": false,
+            "de_novo": false,
+            "unknown_inheritance": false,
+            "is_deleted": 0
+        }
     }
-]
+]

gene2phenotype_project/gene2phenotype_app/fixtures/lgd_variant_type_comment.json

@@ -22,5 +22,17 @@
             "user": 1,
             "date": "2025-05-03T11:38:52Z"
         }
+    },
+    {
+        "model": "gene2phenotype_app.lgdvarianttypecomment",
+        "pk": 3,
+        "fields": {
+            "lgd_variant_type": 6,
+            "comment": "Comment only on the source record's variant type",
+            "is_public": 0,
+            "is_deleted": 0,
+            "user": 1,
+            "date": "2025-06-01T09:00:00Z"
+        }
     }
 ]

gene2phenotype_project/gene2phenotype_app/fixtures/lgd_variant_type_publication.json

@@ -0,0 +1,56 @@
+[
+    {
+        "model": "gene2phenotype_app.lgdvarianttypepublication",
+        "pk": 1,
+        "fields": {
+            "lgd_variant_type": 1,
+            "publication": 2,
+            "is_deleted": 0
+        }
+    },
+    {
+        "model": "gene2phenotype_app.lgdvarianttypepublication",
+        "pk": 2,
+        "fields": {
+            "lgd_variant_type": 2,
+            "publication": 2,
+            "is_deleted": 0
+        }
+    },
+    {
+        "model": "gene2phenotype_app.lgdvarianttypepublication",
+        "pk": 3,
+        "fields": {
+            "lgd_variant_type": 3,
+            "publication": 1,
+            "is_deleted": 0
+        }
+    },
+    {
+        "model": "gene2phenotype_app.lgdvarianttypepublication",
+        "pk": 4,
+        "fields": {
+            "lgd_variant_type": 4,
+            "publication": 2,
+            "is_deleted": 1
+        }
+    },
+    {
+        "model": "gene2phenotype_app.lgdvarianttypepublication",
+        "pk": 5,
+        "fields": {
+            "lgd_variant_type": 5,
+            "publication": 2,
+            "is_deleted": 0
+        }
+    },
+    {
+        "model": "gene2phenotype_app.lgdvarianttypepublication",
+        "pk": 6,
+        "fields": {
+            "lgd_variant_type": 6,
+            "publication": 1,
+            "is_deleted": 0
+        }
+    }
+]

gene2phenotype_project/gene2phenotype_app/fixtures/locus_genotype_disease.json

@@ -151,5 +151,39 @@
             "is_reviewed": 1,
             "is_deleted": 0
         }
+    },
+    {
+        "model": "gene2phenotype_app.locusgenotypedisease",
+        "pk": 10,
+        "fields": {
+            "stable_id": 12,
+            "locus": 1,
+            "genotype": 9,
+            "disease": 6,
+            "mechanism": 1,
+            "mechanism_support": 17,
+            "confidence": 1,
+            "confidence_support": null,
+            "date_review": "2025-06-01T09:00:00Z",
+            "is_reviewed": 1,
+            "is_deleted": 0
+        }
+    },
+    {
+        "model": "gene2phenotype_app.locusgenotypedisease",
+        "pk": 11,
+        "fields": {
+            "stable_id": 13,
+            "locus": 1,
+            "genotype": 9,
+            "disease": 1,
+            "mechanism": 1,
+            "mechanism_support": 17,
+            "confidence": 1,
+            "confidence_support": null,
+            "date_review": "2025-06-01T09:00:00Z",
+            "is_reviewed": 1,
+            "is_deleted": 0
+        }
     }
 ]

gene2phenotype_project/gene2phenotype_app/management/commands/datachecks/__init__.py

@@ -14,6 +14,6 @@
 
 from .MinedPublications import check_mined_publication_status
 
-from .SimilarRecords import get_similar_records
+from .SimilarRecords import get_similar_records, get_records_with_publication_overlap
 
 from .DeletedRecords import check_deleted_records

gene2phenotype_project/gene2phenotype_app/migrations/0021_add_lgd_variant_type_publication.py

@@ -0,0 +1,50 @@
+# Generated by Django 5.2.14 on 2026-07-13 15:43
+
+import django.db.models.deletion
+import simple_history.models
+from django.conf import settings
+from django.db import migrations, models
+
+
+class Migration(migrations.Migration):
+
+    dependencies = [
+        ('gene2phenotype_app', '0020_remove_uniprotannotation_uniprot_ann_uniprot_c49c17_idx_and_more'),
+    ]
+
+    operations = [
+        migrations.CreateModel(
+            name='HistoricalLGDVariantTypePublication',
+            fields=[
+                ('id', models.IntegerField(blank=True, db_index=True)),
+                ('is_deleted', models.SmallIntegerField(default=False)),
+                ('history_id', models.AutoField(primary_key=True, serialize=False)),
+                ('history_date', models.DateTimeField(db_index=True)),
+                ('history_change_reason', models.CharField(max_length=100, null=True)),
+                ('history_type', models.CharField(choices=[('+', 'Created'), ('~', 'Changed'), ('-', 'Deleted')], max_length=1)),
+                ('history_user', models.ForeignKey(null=True, on_delete=django.db.models.deletion.SET_NULL, related_name='+', to=settings.AUTH_USER_MODEL)),
+                ('lgd_variant_type', models.ForeignKey(blank=True, db_constraint=False, null=True, on_delete=django.db.models.deletion.DO_NOTHING, related_name='+', to='gene2phenotype_app.lgdvarianttype')),
+                ('publication', models.ForeignKey(blank=True, db_constraint=False, null=True, on_delete=django.db.models.deletion.DO_NOTHING, related_name='+', to='gene2phenotype_app.publication')),
+            ],
+            options={
+                'verbose_name': 'historical lgd variant type publication',
+                'verbose_name_plural': 'historical lgd variant type publications',
+                'ordering': ('-history_date', '-history_id'),
+                'get_latest_by': ('history_date', 'history_id'),
+            },
+            bases=(simple_history.models.HistoricalChanges, models.Model),
+        ),
+        migrations.CreateModel(
+            name='LGDVariantTypePublication',
+            fields=[
+                ('id', models.AutoField(primary_key=True, serialize=False)),
+                ('is_deleted', models.SmallIntegerField(default=False)),
+                ('lgd_variant_type', models.ForeignKey(on_delete=django.db.models.deletion.PROTECT, related_name='publications', to='gene2phenotype_app.lgdvarianttype')),
+                ('publication', models.ForeignKey(on_delete=django.db.models.deletion.PROTECT, to='gene2phenotype_app.publication')),
+            ],
+            options={
+                'db_table': 'lgd_variant_type_publication',
+                'unique_together': {('lgd_variant_type', 'publication')},
+            },
+        ),
+    ]

gene2phenotype_project/gene2phenotype_app/migrations/0022_populate_lgd_variant_type_publication.py

@@ -0,0 +1,69 @@
+from collections import defaultdict
+
+from django.db import migrations
+
+
+def populate_and_deduplicate(apps, schema_editor):
+    """
+    LGDVariantType used to have a direct (nullable) publication FK, so the same
+    variant type got a separate row per supporting publication.
+    For each (lgd, variant_type_ot) group of rows:
+      - keep the earliest row as canonical
+      - copy every row's publication link into the new LGDVariantTypePublication join table
+      - re-point any comments from the surplus duplicate rows onto the canonical row
+      - merge the inherited/de_novo/unknown_inheritance flags (logical OR), matching the
+        merge semantics the API already applied when displaying these duplicates
+      - delete the surplus duplicate rows
+    """
+    LGDVariantType = apps.get_model("gene2phenotype_app", "LGDVariantType")
+    LGDVariantTypePublication = apps.get_model(
+        "gene2phenotype_app", "LGDVariantTypePublication"
+    )
+    LGDVariantTypeComment = apps.get_model(
+        "gene2phenotype_app", "LGDVariantTypeComment"
+    )
+
+    groups = defaultdict(list)
+    for row in LGDVariantType.objects.all().order_by("id"):
+        groups[(row.lgd_id, row.variant_type_ot_id)].append(row)
+
+    for rows in groups.values():
+        canonical = rows[0]
+
+        for row in rows:
+            if row.publication_id:
+                LGDVariantTypePublication.objects.get_or_create(
+                    lgd_variant_type_id=canonical.id,
+                    publication_id=row.publication_id,
+                    defaults={"is_deleted": row.is_deleted},
+                )
+
+        for row in rows[1:]:
+            LGDVariantTypeComment.objects.filter(lgd_variant_type_id=row.id).update(
+                lgd_variant_type_id=canonical.id
+            )
+            canonical.inherited = canonical.inherited or row.inherited
+            canonical.de_novo = canonical.de_novo or row.de_novo
+            canonical.unknown_inheritance = (
+                canonical.unknown_inheritance or row.unknown_inheritance
+            )
+            if row.is_deleted == 0:
+                canonical.is_deleted = 0
+            row.delete()
+
+        canonical.save()
+
+
+def noop_reverse(apps, schema_editor):
+    pass
+
+
+class Migration(migrations.Migration):
+
+    dependencies = [
+        ("gene2phenotype_app", "0021_add_lgd_variant_type_publication"),
+    ]
+
+    operations = [
+        migrations.RunPython(populate_and_deduplicate, noop_reverse),
+    ]

gene2phenotype_project/gene2phenotype_app/migrations/0023_remove_lgdvarianttype_publication.py

@@ -0,0 +1,25 @@
+# Generated by Django 5.2.14 on 2026-07-13 15:43
+
+from django.db import migrations
+
+
+class Migration(migrations.Migration):
+
+    dependencies = [
+        ('gene2phenotype_app', '0022_populate_lgd_variant_type_publication'),
+    ]
+
+    operations = [
+        migrations.RemoveField(
+            model_name='historicallgdvarianttype',
+            name='publication',
+        ),
+        migrations.AlterUniqueTogether(
+            name='lgdvarianttype',
+            unique_together={('lgd', 'variant_type_ot')},
+        ),
+        migrations.RemoveField(
+            model_name='lgdvarianttype',
+            name='publication',
+        ),
+    ]

gene2phenotype_project/gene2phenotype_app/models.py

@@ -198,10 +198,11 @@ class Meta:
 
 class LGDVariantType(models.Model):
     """
-    Represents the variant type associated with the G2P record and a publication.
+    Represents the variant type associated with the G2P record.
 
-    Types of variants reported in the publication: missense_variant, frameshift_variant, stop_gained, etc.
+    Types of variants reported in publications: missense_variant, frameshift_variant, stop_gained, etc.
     Sequence ontology terms are used to describe variant types.
+    A variant type can be supported by one or more publications - see LGDVariantTypePublication.
     """
 
     id = models.AutoField(primary_key=True)
@@ -212,19 +213,37 @@ class LGDVariantType(models.Model):
     inherited = models.BooleanField(default=False)
     de_novo = models.BooleanField(default=False)
     unknown_inheritance = models.BooleanField(default=False)
-    publication = models.ForeignKey("Publication", on_delete=models.PROTECT, null=True)
     is_deleted = models.SmallIntegerField(null=False, default=False)
     history = HistoricalRecords()
 
     class Meta:
         db_table = "lgd_variant_type"
-        unique_together = ["lgd", "variant_type_ot", "publication"]
+        unique_together = ["lgd", "variant_type_ot"]
         indexes = [
             models.Index(fields=["lgd", "variant_type_ot"]),
             models.Index(fields=["variant_type_ot"]),
         ]
 
 
+class LGDVariantTypePublication(models.Model):
+    """
+    Links a LGDVariantType to the publication(s) that support it.
+    A single variant type can be evidenced by more than one publication.
+    """
+
+    id = models.AutoField(primary_key=True)
+    lgd_variant_type = models.ForeignKey(
+        "LGDVariantType", related_name="publications", on_delete=models.PROTECT
+    )
+    publication = models.ForeignKey("Publication", on_delete=models.PROTECT)
+    is_deleted = models.SmallIntegerField(null=False, default=False)
+    history = HistoricalRecords()
+
+    class Meta:
+        db_table = "lgd_variant_type_publication"
+        unique_together = ["lgd_variant_type", "publication"]
+
+
 class LGDVariantTypeDescription(models.Model):
     """
     Represents the HGVS description linked to the LGD record and the publication.