Merge branch 'master' into numpy-rng

Author: jonbrenas

malariagen_data/anoph/snp_data.py

@@ -1872,9 +1872,10 @@ def biallelic_snp_calls(
 
                 # Apply missingness condition.
                 if max_missing_an is not None:
-                    an_missing = (ds_out.sizes["samples"] * ds_out.sizes["ploidy"]) - an
+                    an_total = ds_out.sizes["samples"] * ds_out.sizes["ploidy"]
+                    an_missing = an_total - an
                     if isinstance(max_missing_an, float):
-                        an_missing_frac = an_missing / an
+                        an_missing_frac = an_missing / an_total
                         loc_missing = an_missing_frac <= max_missing_an
                     else:
                         loc_missing = an_missing <= max_missing_an

malariagen_data/anopheles.py

@@ -1066,7 +1066,7 @@ def cohort_diversity_stats(
             cohort_label, cohort_query = cohort
 
         else:
-            raise TypeError(r"invalid cohort parameter: {cohort!r}")
+            raise TypeError(f"invalid cohort parameter: {cohort!r}")
 
         debug("access allele counts")
         ac = self.snp_allele_counts(

malariagen_data/util.py

@@ -430,6 +430,7 @@ def _init_filesystem(url, **kwargs):
     """Initialise a fsspec filesystem from a given base URL and parameters."""
 
     storage_options = None  # To prevent using before assignment (Pylint).
+    simplecache_options = kwargs.pop("simplecache", None)
 
     # Special case Google Cloud Storage, authenticate the user.
     if "gs://" in url or "gcs://" in url:
@@ -487,6 +488,9 @@ def _init_filesystem(url, **kwargs):
         # Some other kind of URL, pass through kwargs as-is.
         storage_options = kwargs
 
+    if simplecache_options is not None:
+        storage_options["simplecache"] = simplecache_options
+
     # Process the URL using fsspec.
     fs, path = url_to_fs(url, **storage_options)
 

tests/anoph/test_snp_data.py

@@ -1723,8 +1723,9 @@ def test_biallelic_snp_calls_and_diplotypes_with_conditions_fractional(
     an = ac.sum(axis=1)
     ac_min = ac.min(axis=1)
     assert np.all((ac_min / an) >= min_minor_ac)
-    an_missing = (ds.sizes["samples"] * ds.sizes["ploidy"]) - an
-    assert np.all((an_missing / an) <= max_missing_an)
+    an_total = ds.sizes["samples"] * ds.sizes["ploidy"]
+    an_missing = an_total - an
+    assert np.all((an_missing / an_total) <= max_missing_an)
     gt = ds["call_genotype"].values
     ac_check = allel.GenotypeArray(gt).count_alleles(max_allele=1)
     assert np.all(ac == ac_check)