Fix #1108: Replace truthiness checks with explicit None checks in genome sequence slicing

PlasmodiumDataResource._subset_genome_sequence_region() and
AnophelesGenomeSequenceData.genome_sequence() used truthiness checks
(if region.start:) which treated 0 as falsy, silently skipping slicing.

Changes:
- plasmodium.py: Replace truthiness with 'is not None', add start/end >= 1
  validation, fix docstring typo 'Sebset' -> 'Subset'
- genome_sequence.py: Same truthiness fix + bounds validation
- snp_data.py: Fix same bug in is_accessible() (line 1726)
- test_plasmodium.py: Add 3 regression tests for valid slice, start=0, end=0

The library uses 1-based genomic coordinates (start - 1 conversion), so
start=0 and end=0 now raise ValueError instead of silently misbehaving.

Author: khushthecoder

malariagen_data/anoph/genome_sequence.py

@@ -117,9 +117,13 @@ def genome_sequence(
 
         # Deal with region start and stop.
         slice_start = slice_stop = None
-        if resolved_region.start:
+        if resolved_region.start is not None:
+            if resolved_region.start < 1:
+                raise ValueError("Region start must be >= 1 or None.")
             slice_start = resolved_region.start - 1
-        if resolved_region.end:
+        if resolved_region.end is not None:
+            if resolved_region.end < 1:
+                raise ValueError("Region end must be >= 1 or None.")
             slice_stop = resolved_region.end
         loc_region = slice(slice_start, slice_stop)
 

malariagen_data/anoph/snp_data.py

@@ -1723,7 +1723,7 @@ def is_accessible(
 
         # Access SNP site positions.
         pos = self.snp_sites(region=resolved_region, field="POS").compute()
-        if resolved_region.start:
+        if resolved_region.start is not None:
             offset = resolved_region.start
         else:
             offset = 1

malariagen_data/plasmodium.py

@@ -232,17 +232,22 @@ def _resolve_region(self, region):
     def _subset_genome_sequence_region(
         self, genome, region, inline_array=True, chunks="native"
     ):
-        """Sebset reference genome sequence."""
+        """Subset reference genome sequence."""
         region = self._resolve_region(region)
         z = genome[region.contig]
 
         d = _da_from_zarr(z, inline_array=inline_array, chunks=chunks)
 
-        if region.start:
+        if region.start is not None:
+            if region.start < 1:
+                raise ValueError("Region start must be >= 1 or None.")
             slice_start = region.start - 1
         else:
             slice_start = None
-        if region.end:
+            
+        if region.end is not None:
+            if region.end < 1:
+                raise ValueError("Region end must be >= 1 or None.")
             slice_stop = region.end
         else:
             slice_stop = None

tests/test_plasmodium.py

@@ -690,6 +690,34 @@ def test_subset_genome_sequence_contig2(self):
         assert sequence.shape == (15,)
         assert sequence.ndim == 1
 
+    def test_subset_genome_sequence_region_slice(self):
+        from malariagen_data.util import Region
+        # contig1 is np.arange(10)
+        # Valid slice 1-3 -> indices 0,1,2 (length 3)
+        region = Region("contig1", 1, 3)
+        sequence = self.test_plasmodium_class._subset_genome_sequence_region(
+            genome=self.test_ref_zarr, region=region
+        )
+        assert isinstance(sequence, da.Array)
+        assert sequence.shape == (3,)
+        assert sequence.compute().tolist() == [0, 1, 2]
+
+    def test_subset_genome_sequence_region_invalid_start_0(self):
+        from malariagen_data.util import Region
+        region = Region("contig1", 0, 3)
+        with self.assertRaises(ValueError):
+            self.test_plasmodium_class._subset_genome_sequence_region(
+                genome=self.test_ref_zarr, region=region
+            )
+
+    def test_subset_genome_sequence_region_invalid_end_0(self):
+        from malariagen_data.util import Region
+        region = Region("contig1", 1, 0)
+        with self.assertRaises(ValueError):
+            self.test_plasmodium_class._subset_genome_sequence_region(
+                genome=self.test_ref_zarr, region=region
+            )
+
 
 if __name__ == "__main__":
     unittest.main()