Merge pull request #708 from holtgrewe/fix-hgvs-projection-ins-stop-retained

fix: projection issue for stop retained insertion (#707)

Author: reece

src/hgvs/utils/altseq_to_hgvsp.py

@@ -97,7 +97,8 @@ def build_hgvsp(self):
                     # get size diff from diff in ref/alt lengths
                     start = self._alt_data.variant_start_aa - 1
                     delta = len(self._alt_seq) - len(self._ref_seq)
-                    while self._ref_seq[start] == self._alt_seq[start]:
+                    start_max = min(len(self._ref_seq), len(self._alt_seq))
+                    while start < start_max and self._ref_seq[start] == self._alt_seq[start]:
                         start += 1
                     offset = start + abs(delta)
 
@@ -190,10 +191,15 @@ def _convert_to_sequence_variants(self, variant, acc):
             self._is_ambiguous = True  # side-effect
 
         if insertion and insertion.find("*") == 0:  # stop codon at variant position
-            aa_start = aa_end = AAPosition(base=start, aa=deletion[0])
-            ref = ""
-            alt = "*"
             is_sub = True
+            alt = "*"
+            if start == len(self._alt_seq) and not deletion:
+                # insertion of stop codon at end of sequence, a special case of "stop_retained"
+                aa_start = aa_end = AAPosition(base=start, aa="*")
+                ref = "*"
+            else:
+                aa_start = aa_end = AAPosition(base=start, aa=deletion[0])
+                ref = ""
 
         elif start == len(self._ref_seq):  # extension
             if self._alt_seq[-1] == "*":

tests/data/cache-py3.hdp

@@ -37,6 +37,13 @@ def setUpClass(cls):
         cls.vm = hgvs.variantmapper.VariantMapper(cls.hdp)
         cls.hp = hgvs.parser.Parser()
 
+    def test_map_stop_retained(self):
+        hgvs_c = "NM_001253909.2:c.416_417insGTG"
+        var_c = self.hp.parse_hgvs_variant(hgvs_c)
+        var_p = self.vm.c_to_p(var_c)
+
+        assert str(var_p) == "NP_001240838.1:p.(Ter140Ter)"
+
     def test_gcrp_invalid_input_type(self):
         hgvs_g = "NC_000007.13:g.36561662C>T"
         hgvs_c = "NM_001637.3:c.1582G>A"