Handle TIDDIT style SV type names

CHANGELOG.md

@@ -13,6 +13,7 @@ About changelog [here](https://keepachangelog.com/en/1.0.0/)
 ### Fixed
 - Restore command accepts custom database name
 - Restore command uses either database URI or host:port params
+- Tiddit style variants (IDUP, TDUP or DUP:TANDEM also in SVTYPE, not only ALT field) collapse to primary type (DUP, DEL) 
 
 ## [2.5.2]
 

loqusdb/build_models/variant.py

@@ -98,6 +98,11 @@ def get_coords(variant):
     coordinates["end"] = end
 
     sv_type = variant.INFO.get("SVTYPE")
+    if sv_type:
+        sv_type = sv_type.split(":")[0]
+    if sv_type in ["TDUP", "IDUP"]:
+        sv_type = "DUP"
+
     length = variant.INFO.get("SVLEN")
     sv_len = abs(length) if length else end - pos
     # Translocations will sometimes have a end chrom that differs from chrom

tests/conftest.py

@@ -424,6 +424,32 @@ def duptandem_variant(request):
         info_dict={"END": 3092849, "SVLEN": 223, "SVTYPE": "DUP"},
     )
 
+@pytest.fixture(scope="function")
+def duptandem_tiddit_variant(request):
+    return CyvcfVariant(
+        chrom="1",
+        ref="A",
+        alt="<DUP:TANDEM>",
+        pos=3092626,
+        end=3092849,
+        var_type="sv",
+        info_dict={"END": 3092849, "SVLEN": 223, "SVTYPE": "DUP:TANDEM"},
+    )
+
+
+@pytest.fixture(scope="function")
+def duptandem_old_tiddit_variant(request):
+    return CyvcfVariant(
+        chrom="1",
+        ref="A",
+        alt="<TDUP>",
+        pos=3092626,
+        end=3092849,
+        var_type="sv",
+        info_dict={"END": 3092849, "SVLEN": 223, "SVTYPE": "TDUP"},
+    )
+
+
 
 @pytest.fixture(scope="function")
 def translocation_variant(request):

tests/vcf_tools/test_format_sv_variant.py

@@ -82,6 +82,44 @@ def test_format_dup_tandem(duptandem_variant, case_obj):
     assert formated_variant["alt"] == variant.ALT[0]
     assert formated_variant["sv_type"] == "DUP"
 
+def test_format_tiddit_dup_tandem(duptandem_tiddit_variant, case_obj):
+    ## GIVEN a TIDDIT style notation, ie both SVTYPE info and alt field tag has subtype, eg DUP:TANDEM
+    variant = duptandem_tiddit_variant
+    case_id = case_obj["case_id"]
+    ## WHEN parsing the variant
+    formated_variant = build_variant(variant=variant, case_obj=case_obj, case_id=case_id)
+
+    ## THEN assert the sv is parsed correct
+    assert formated_variant["chrom"] == variant.CHROM
+    assert formated_variant["end_chrom"] == variant.CHROM
+    assert formated_variant["pos"] == variant.POS
+    assert formated_variant["end"] == variant.INFO["END"]
+    assert formated_variant["sv_len"] == abs(variant.INFO["SVLEN"])
+
+    assert formated_variant["ref"] == variant.REF
+    assert formated_variant["alt"] == variant.ALT[0]
+    assert formated_variant["sv_type"] == "DUP"
+
+
+def test_format_old_tiddit_dup_tandem(duptandem_old_tiddit_variant, case_obj):
+    ## GIVEN a dup with older TIDDIT notation (TDUP, IDUP, ...)
+    variant = duptandem_old_tiddit_variant
+    case_id = case_obj["case_id"]
+    ## WHEN parsing the variant
+    formated_variant = build_variant(variant=variant, case_obj=case_obj, case_id=case_id)
+
+    ## THEN assert the sv is parsed correct
+    assert formated_variant["chrom"] == variant.CHROM
+    assert formated_variant["end_chrom"] == variant.CHROM
+    assert formated_variant["pos"] == variant.POS
+    assert formated_variant["end"] == variant.INFO["END"]
+    assert formated_variant["sv_len"] == abs(variant.INFO["SVLEN"])
+
+    assert formated_variant["ref"] == variant.REF
+    assert formated_variant["alt"] == variant.ALT[0]
+    assert formated_variant["sv_type"] == "DUP"
+
+
 
 def test_format_translocation(translocation_variant, case_obj):
     ## GIVEN a small insertion (This means that the insertion is included in ALT field)