Account for scaffolds with gaps at their ends

Author: jgrg

pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "tola-agp-tpf-utils"
-version = "1.3.2"
+version = "1.3.3"
 readme = "README.md"
 requires-python = ">=3.11"
 dependencies = [

src/tola/assembly/indexed_assembly.py

@@ -74,10 +74,16 @@ def find_overlaps(self, bait: Fragment) -> OverlapResult | None:
             return None
         i_ovr, j_ovr = rng
 
-        # Walk start and end pointers back to ignore Gaps on the ends
+        # Walk start and end pointers inwards to ignore any Gaps on the ends
+        # of the match.
+        last_i = len(scffld.rows) - 1
         while isinstance(scffld.rows[i_ovr], Gap):
+            if i_ovr == last_i:
+                return None
             i_ovr += 1
         while isinstance(scffld.rows[j_ovr], Gap):
+            if j_ovr == 0:
+                return None
             j_ovr -= 1
         if not i_ovr <= j_ovr:
             return None

tests/indexed_assembly_test.py

@@ -54,6 +54,45 @@ def test_create():
         x = i3.find_overlaps(Fragment("scaffold_1", 1, 100, 1))
 
 
+def test_gaps_at_scaffold_ends():
+    frag = Fragment("ctg_1", 1, 1000, 1)
+    ia1 = IndexedAssembly(
+        "gaps at ends",
+        scaffolds=[
+            Scaffold(
+                "scffld_1",
+                rows=[
+                    Gap(200, "scaffold"),
+                    frag,
+                    Gap(50, "scaffold"),
+                ],
+            ),
+        ],
+    )
+    found1 = ia1.find_overlaps(Fragment("scffld_1", 1, 1250, 1))
+    assert found1.rows == [frag]
+
+    found2 = ia1.find_overlaps(Fragment("scffld_1", 1, 2, 1))
+    assert found2 is None
+
+    ia2 = IndexedAssembly(
+        "all gaps",
+        scaffolds=[
+            Scaffold(
+                "scffld_1",
+                rows=[
+                    Gap(200, "scaffold"),
+                    Gap(100, "scaffold"),
+                    Gap(50, "scaffold"),
+                ],
+            ),
+        ],
+    )
+
+    found3 = ia2.find_overlaps(Fragment("scffld_1", 1, 350, 1))
+    assert found3 is None
+
+
 def test_find_overlapping():
     """
     IndexedAssembly: Random assembly