add option for delins or two substitutions (#4)

* prefer two substitions rather than delins in codons * change default for prefer_delins
VariantEffect · Nov 26, 2021 · 847a782 · 847a782
1 parent 4aac65f
commit 847a782
Show file tree

Hide file tree

Showing 2 changed files with 32 additions and 10 deletions.
diff --git a/mavetools/convert/convert.py b/mavetools/convert/convert.py
@@ -9,6 +9,7 @@ def codon_sub_to_mavehgvs(
     variant_codon: str,
     aa_position: int,
     target_id: Optional[str] = None,
+    prefer_delins: bool = True,
 ) -> str:
     """Create a MAVE-HGVS coding variant string describing the change between two codons.
 
@@ -23,6 +24,11 @@ def codon_sub_to_mavehgvs(
         This will be used to calculate the nucleotide positions.
     target_id : Optional[str]
         Optional target identifier for the resulting variant.
+        Default ``None``.
+    prefer_delins : bool
+        If True, consecutive two-base changes will be described as a deletion-insertion;
+        otherwise they will be described as two single-nucleotide substitutions.
+        Default ``True``.
 
     Returns
     -------
@@ -64,13 +70,25 @@ def codon_sub_to_mavehgvs(
                     f"{variant_pos + 2}{target_codon[2]}>{variant_codon[2]}]"
                 )
             elif changes[0]:  # delins of first two bases
-                variant_string = (
-                    f"c.{variant_pos}_{variant_pos + 1}delins{variant_codon[:2]}"
-                )
+                if prefer_delins:
+                    variant_string = (
+                        f"c.{variant_pos}_{variant_pos + 1}delins{variant_codon[:2]}"
+                    )
+                else:
+                    variant_string = (
+                        f"c.[{variant_pos}{target_codon[0]}>{variant_codon[0]};"
+                        f"{variant_pos + 1}{target_codon[1]}>{variant_codon[1]}]"
+                    )
             else:  # delins of last two bases
-                variant_string = (
-                    f"c.{variant_pos + 1}_{variant_pos + 2}delins{variant_codon[1:]}"
-                )
+                if prefer_delins:
+                    variant_string = (
+                        f"c.{variant_pos + 1}_{variant_pos + 2}delins{variant_codon[1:]}"
+                    )
+                else:
+                    variant_string = (
+                        f"c.[{variant_pos + 1}{target_codon[1]}>{variant_codon[1]};"
+                        f"{variant_pos + 2}{target_codon[2]}>{variant_codon[2]}]"
+                    )
         elif sum(changes) == 3:  # full codon delins
             variant_string = f"c.{variant_pos}_{variant_pos + 2}delins{variant_codon}"
         else:  # pragma: nocover

diff --git a/tests/test_convert/test_convert.py b/tests/test_convert/test_convert.py
@@ -12,11 +12,15 @@ def test_valid_substitution(self):
             (("ATG", "AAG", 11), "c.32T>A"),
             (("ATG", "ATT", 88), "c.264G>T"),
             (("ATG", "TTT", 1), "c.[1A>T;3G>T]"),
-            (("ATG", "TAG", 1), "c.1_2delinsTA"),
-            (("ATG", "AGT", 1), "c.2_3delinsGT"),
+            (("ATG", "TAG", 1, None, False), "c.[1A>T;2T>A]"),
+            (("ATG", "AGT", 1, None, False), "c.[2T>G;3G>T]"),
+            (("ATG", "TAG", 1, None, True), "c.1_2delinsTA"),
+            (("ATG", "AGT", 1, None, True), "c.2_3delinsGT"),
             (("ATG", "TTT", 2), "c.[4A>T;6G>T]"),
-            (("ATG", "TAG", 11), "c.31_32delinsTA"),
-            (("ATG", "AGT", 88), "c.263_264delinsGT"),
+            (("ATG", "TAG", 11, None, False), "c.[31A>T;32T>A]"),
+            (("ATG", "AGT", 88, None, False), "c.[263T>G;264G>T]"),
+            (("ATG", "TAG", 11, None, True), "c.31_32delinsTA"),
+            (("ATG", "AGT", 88, None, True), "c.263_264delinsGT"),
             (("ATG", "CAT", 1), "c.1_3delinsCAT"),
             (("ATG", "CAT", 101), "c.301_303delinsCAT"),
         ]