Merge pull request #337 from bcgsc/bugfix/MAVIS_summary_clustering

Bugfix to Summary module clustering output

.github/workflows/build.yml

@@ -96,6 +96,7 @@ jobs:
         run: |
           python -m pip install --upgrade pip setuptools wheel
           pip install mavis_config pandas snakemake
+          pip install tabulate==0.8.9
       - uses: eWaterCycle/setup-singularity@v6
         with:
           singularity-version: 3.6.4

setup.cfg

@@ -45,6 +45,7 @@ install_requires =
     Shapely>=1.6.4.post1
     shortuuid>=0.5.0
     svgwrite
+    tabulate==0.8.9
     typing_extensions>=4
 setup_requires =
     pip>=9.0.0

src/mavis/summary/summary.py

@@ -210,10 +210,11 @@ def annotate_dgv(bpps, dgv_regions_by_reference_name, distance=0):
     for bpp in [
         b for b in bpps if not b.interchromosomal and b.break1.chr in dgv_regions_by_reference_name
     ]:
+        bpp.data['dgv'] = []
         for dgv_region in dgv_regions_by_reference_name[bpp.break1.chr]:
             dist = abs(Interval.dist(Interval(dgv_region.start), bpp.break1))
             if dist > lowest_resolution + distance:
-                break
+                continue
             elif (
                 dist > distance
                 or abs(Interval.dist(Interval(dgv_region.end), bpp.break2)) > distance
@@ -224,9 +225,10 @@ def annotate_dgv(bpps, dgv_regions_by_reference_name, distance=0):
                 refname = dgv_region.reference_object.name
             except AttributeError:
                 pass
-            bpp.data['dgv'] = '{}({}:{}-{})'.format(
-                dgv_region.name, refname, dgv_region.start, dgv_region.end
+            bpp.data['dgv'].append(
+                '{}({}:{}-{})'.format(dgv_region.name, refname, dgv_region.start, dgv_region.end)
             )
+        bpp.data['dgv'] = ';'.join(bpp.data['dgv'])
 
 
 def get_pairing_state(

tests/data/mock_dgv_annotation.txt

@@ -1,6 +1,8 @@
 chr	start	end	name
 1	1	2300000	nsv482937
 1	10001	22118	dgv1n82
+1	10001	22120	rgv2n98
+1	10001	22221	rgv2n99
 1	10001	127330	nsv7879
 1	10191	10281	nsv958854
 1	10377	177417	nsv428112

tests/test_mavis/summary/test_summary.py

@@ -1,9 +1,10 @@
 import pytest
 from mavis.breakpoint import Breakpoint, BreakpointPair
 from mavis.constants import CALL_METHOD, COLUMNS, PROTOCOL, STRAND, SVTYPE
-from mavis.summary.summary import filter_by_annotations
+from mavis.summary.summary import filter_by_annotations, annotate_dgv
+from mavis.annotate.file_io import load_masking_regions
 
-from ...util import todo
+from ...util import todo, get_data
 
 
 @pytest.fixture
@@ -40,11 +41,34 @@ def genomic_event2():
     )
 
 
+@pytest.fixture
+def genomic_event3():
+    return BreakpointPair(
+        Breakpoint('1', 10001),
+        Breakpoint('1', 22118),
+        opposing_strands=True,
+        **{
+            COLUMNS.event_type: SVTYPE.DEL,
+            COLUMNS.call_method: CALL_METHOD.CONTIG,
+            COLUMNS.fusion_sequence_fasta_id: None,
+            COLUMNS.protocol: PROTOCOL.GENOME,
+            COLUMNS.fusion_cdna_coding_start: None,
+            COLUMNS.fusion_cdna_coding_end: None,
+            COLUMNS.tracking_id: "genomic_event3",
+        }
+    )
+
+
 @pytest.fixture
 def best_transcripts():
     return {'ABCA': True, 'ABCD': True}
 
 
+@pytest.fixture
+def dgv_event2():
+    return load_masking_regions(get_data("mock_dgv_annotation.txt"))
+
+
 class TestFilterByAnnotations:
     def test_filter_by_annotations_two_best_transcripts(
         self, genomic_event1, genomic_event2, best_transcripts
@@ -174,3 +198,10 @@ def test_filtering_events_split(self):
     @todo
     def test_get_pairing_state(self):
         pass
+
+
+class TestFilterByCallMethod:
+    def test_annotate_dgv_distance_bed(self, genomic_event3, dgv_event2):
+        bpps = [genomic_event3]
+        annotate_dgv(bpps, dgv_event2, 103)
+        assert len(bpps[0].data['dgv'].split(';')) == 3