Revert "modiefied demo files, save BND SV variants to database"

This reverts commit 95f8d72.

cgbeacon2/cli/add.py

@@ -52,15 +52,15 @@ def demo(ctx):
     ctx.invoke(
         variants,
         ds=ds_id,
-        vcf="cgbeacon2/resources/demo/test_trio.vcf.gz",
+        vcf="cgbeacon2/resources/demo/643594.clinical.vcf.gz",
         sample=[sample],
     )
 
     # Invoke add variants command to import all SV variants from demo sample
     ctx.invoke(
         variants,
         ds=ds_id,
-        vcf="cgbeacon2/resources/demo/test_trio.SV.vcf.gz",
+        vcf="cgbeacon2/resources/demo/643594.clinical.SV.vcf.gz",
         sample=[sample],
     )
 

cgbeacon2/resources/__init__.py

@@ -1,9 +1,9 @@
 import pkg_resources
 
 ###### Files ######
-test_snv_vcf = "resources/demo/test_trio.vcf.gz"
-test_sv_vcf = "resources/demo/test_trio.SV.vcf.gz"
-empty_vcf = "resources/demo/empty.SV.vcf.gz"
+test_snv_vcf = "resources/demo/643594.clinical.vcf.gz"
+test_sv_vcf = "resources/demo/643594.clinical.SV.vcf.gz"
+empty_vcf = "resources/demo/empty.clinical.SV.vcf.gz"
 panel1 = "resources/demo/panel1.bed"
 panel2 = "resources/demo/panel2.bed"
 

cgbeacon2/utils/add.py

@@ -4,7 +4,7 @@
 
 from cgbeacon2.constants import CHROMOSOMES
 from cgbeacon2.models.variant import Variant
-from cgbeacon2.utils.parse import variant_called, bnd_mate_name
+from cgbeacon2.utils.parse import variant_called
 
 LOG = logging.getLogger(__name__)
 
@@ -78,8 +78,7 @@ def add_variants(database, vcf_obj, samples, assembly, dataset_id, nr_variants):
     inserted_vars = 0
     with Bar("Processing", max=nr_variants) as bar:
         for vcf_variant in vcf_obj:
-            chrom = vcf_variant.CHROM.replace("chr", "")
-            if chrom not in CHROMOSOMES:
+            if vcf_variant.CHROM not in CHROMOSOMES:
                 LOG.warning(
                     f"chromosome '{vcf_variant.CHROM}' not included in canonical chromosome list, skipping it."
                 )
@@ -94,18 +93,15 @@ def add_variants(database, vcf_obj, samples, assembly, dataset_id, nr_variants):
                 continue  # variant was not called in samples of interest
 
             parsed_variant = dict(
-                chromosome=chrom,
+                chromosome=vcf_variant.CHROM,
                 start=vcf_variant.start,  # 0-based coordinate
                 end=vcf_variant.end,  # 0-based coordinate
                 reference_bases=vcf_variant.REF,
                 alternate_bases=vcf_variant.ALT,
             )
 
             if vcf_variant.var_type == "sv":
-                sv_type = vcf_variant.INFO["SVTYPE"]
-                parsed_variant["variant_type"] = sv_type
-                if sv_type == "BND":
-                    parsed_variant["mate_name"] = bnd_mate_name(vcf_variant.ALT, chrom)
+                parsed_variant["variant_type"] = vcf_variant.INFO["SVTYPE"]
 
             else:
                 parsed_variant["variant_type"] = vcf_variant.var_type.upper()

cgbeacon2/utils/parse.py

@@ -2,13 +2,9 @@
 import logging
 from cyvcf2 import VCF
 import os
-import re
 from pybedtools.bedtool import BedTool
 from tempfile import NamedTemporaryFile
 
-BND_ALT_PATTERN = re.compile(r".*[\],\[](.*?):(.*?)[\],\[]")
-CHR_PATTERN = re.compile(r"(chr)?(.*)", re.IGNORECASE)
-
 LOG = logging.getLogger(__name__)
 
 
@@ -32,29 +28,6 @@ def get_vcf_samples(vcf_file):
     return vcf_samples
 
 
-def bnd_mate_name(alt, chrom):
-    """Returns chromosome and mate for a BND variant
-
-    Accepts:
-        alt(str): vcf_variant.ALT
-        chrom(st): cf_variant.CHROM
-
-    Returns:
-        end_chr(str): a chromosome (1-22, X, Y, MT)
-    """
-    end_chrom = chrom
-    if ":" not in alt:
-        return end_chrom
-
-    match = BND_ALT_PATTERN.match(alt)
-    # BND will often be translocations between different chromosomes
-    if match:
-        other_chrom = match.group(1)
-        match = CHR_PATTERN.match(other_chrom)
-        end_chrom = match.group(2)
-    return end_chrom
-
-
 def extract_variants(vcf_file, samples=None, filter=None):
     """Parse a VCF file and return its variants as cyvcf2.VCF objects
 

tests/cli/add/test_add_variants.py

@@ -399,7 +399,7 @@ def test_add_sv_variants(mock_app, public_dataset, database):
     saved_vars = list(database["variant"].find())
     assert len(saved_vars) > 0
 
-    valid_types = ["INS", "DUP", "DEL", "INV", "BND"]
+    valid_types = ["INS", "DUP", "DEL", "INV"]
     # AND all of them should have a valid SV variant type
     for var in saved_vars:
         assert var["variantType"] in valid_types