Merge pull request #562 from uclahs-cds/czhu-fix-call-variant-alt-splice

Fixed alternative splicing deletion that starts at 3rd nucleotide of start codon

CHANGELOG.md

@@ -10,6 +10,10 @@ This project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.htm
 
 ## [Unreleased]
 
+## [0.9.4] - 2022-09-07
+
+- Fixed issue of alternative splicing deletion that starts at the third nucleotide of start codon. Those variants are now skipped. #560
+
 ## [0.9.3] - 2022-08-26
 
 ### Fixed

moPepGen/__init__.py

@@ -5,7 +5,7 @@
 from typing import Iterable, IO
 
 
-__version__ = '0.9.3'
+__version__ = '0.9.4'
 
 ## Error messages
 ERROR_INDEX_IN_INTRON = 'The genomic index seems to be in an intron'

moPepGen/seqvar/VariantRecord.py

@@ -18,6 +18,7 @@
 SINGLE_NUCLEOTIDE_SUBSTITUTION = ['SNV', 'SNP', 'INDEL', 'RNAEditingSite']
 ATTRS_POSITION = ['START', 'DONOR_START', 'ACCEPTER_START', 'ACCEPTER_POSITION']
 ALTERNATIVE_SPLICING_TYPES = ['Insertion', 'Deletion', 'Substitution']
+RMATS_TYPES = ['SE', 'RI', 'A3SS', 'A5SS', 'MXE']
 
 class VariantRecord():
     """ Defines the location, ref and alt of a genomic variant.
@@ -228,6 +229,10 @@ def is_fusion(self) -> bool:
         """ Check if this is a fusion """
         return self.type == 'Fusion'
 
+    def is_alternative_splicing(self) -> bool:
+        """ Check if this is an alternative splicing event """
+        return any(self.id.startswith(x) for x in RMATS_TYPES)
+
     def is_in_frame_fusion(self, anno:GenomicAnnotation):
         """ Check if this is a in-frame fusion. A in-frame fusion is only when
         both donor and accepter transcripts a protein coding (which known

moPepGen/svgraph/ThreeFrameTVG.py

@@ -905,7 +905,8 @@ def create_variant_graph(self, variants:List[seqvar.VariantRecord],
 
             if variant.location.start == start_index - 1 \
                     and (variant.is_insertion() or variant.is_deletion()) \
-                    and not variant.is_fusion():
+                    and not variant.is_fusion() \
+                    and not variant.is_alternative_splicing():
                 variant.to_end_inclusion(self.seq)
 
             # Skip variants that the position is smaller than the first NT

test/unit/test_three_frame_tvg.py

@@ -432,6 +432,29 @@ def test_create_variant_graph_frameshifting(self):
         expected = {'AAA', 'AA', 'A'}
         self.assertEqual(received, expected)
 
+    def test_create_variant_graph_deletion_alt_splice(self):
+        """ alternative splicing deletion that starts with the third nucleotide
+        of start codon should be skipped. """
+        data = {
+            1: ['AAATAAATAAAT', ['RF0'], [], 0],
+            2: ['AATAAATAAAT',  ['RF1'], [], 1],
+            3: ['ATAAATAAAT',   ['RF2'], [], 2]
+        }
+
+        var_data = [
+            (3, 10, 'T', '<DEL>', 'Deletion', 'SE-10')
+        ]
+
+        graph, _ = create_three_frame_tvg(data, 'AAATAAATAAAT')
+        graph.seq.orf = FeatureLocation(start=1, end=4)
+        graph.has_known_orf = True
+        variants = create_variants(var_data)
+        graph.create_variant_graph(variants, None, None, None)
+        received = {str(list(n.out_edges)[0].out_node.seq.seq) \
+            for n in graph.reading_frames}
+        expected = {x[0] for x in data.values()}
+        self.assertEqual(received, expected)
+
     def test_apply_insertion_case1(self):
         """ apply_insertion """
         anno = create_genomic_annotation(ANNOTATION_DATA)