Merge pull request #937 from 31puneet/fix/g123-empty-array-guard

fix: add guard for empty array in g123 when window_size exceeds avail

Author: jonbrenas

malariagen_data/anoph/g123.py

@@ -133,6 +133,13 @@ def _g123_gwss(
             chunks=chunks,
         )
 
+        if gt.shape[0] < window_size:
+            raise ValueError(
+                f"Not enough sites ({gt.shape[0]}) for window size "
+                f"({window_size}). Please reduce the window size or "
+                f"use different site selection criteria."
+            )
+
         with self._spinner("Compute G123"):
             g123 = allel.moving_statistic(gt, statistic=_garud_g123, size=window_size)
             x = allel.moving_statistic(pos, statistic=np.mean, size=window_size)
@@ -240,6 +247,12 @@ def _g123_calibration(
 
         calibration_runs: Dict[str, np.ndarray] = dict()
         for window_size in self._progress(window_sizes, desc="Compute G123"):
+            if gt.shape[0] < window_size:
+                raise ValueError(
+                    f"Not enough sites ({gt.shape[0]}) for window size "
+                    f"({window_size}). Please reduce the window size or "
+                    f"use different site selection criteria."
+                )
             g123 = allel.moving_statistic(gt, statistic=_garud_g123, size=window_size)
             calibration_runs[str(window_size)] = g123
 

tests/anoph/conftest.py

@@ -2260,7 +2260,7 @@ def init_hap_sites(self):
             path=path,
             contigs=self.contigs,
             snp_sites=self.snp_sites,
-            p_site=np.random.random(),
+            p_site=np.random.uniform(0.5, 1.0),
         )
 
     def init_haplotypes(self):