fix: out of bounds issue on 5'-to-3' shifting beyond CDS (#94)

src/mapper/altseq.rs

@@ -777,6 +777,17 @@ impl AltSeqToHgvsp {
             is_frameshift,
         } = &record;
 
+        // Handle the case of the variant being after the end of the protein sequence.  This can
+        // happen when the variant is 5'-to-3' shifted beyond the stop codon.  In this case, we
+        // can simply return `ProtLocEdit::NoChange` which will display as `p.=`.
+        if *start as usize > self.ref_seq().len() {
+            return Ok(HgvsVariant::ProtVariant {
+                accession: Accession::new(protein_accession),
+                gene_symbol: None,
+                loc_edit: ProtLocEdit::NoChange,
+            });
+        }
+
         // defaults
         let mut is_dup = false; // assume no dup
         let mut fsext_len = UncertainLengthChange::default(); // fs or ext length

src/normalizer.rs

@@ -413,7 +413,7 @@ impl<'a> Normalizer<'a> {
             if stop < ref_seq.len().try_into()? || start == orig_start {
                 break;
             }
-            // If stop at the end of the window, try to extend teh shuffling to the right.
+            // If stop at the end of the window, try to extend the shuffling to the right.
             base += start - orig_start;
             stop -= start - orig_start;
             start = orig_start;