Fix a side effect of to_plink, version 0.9.1

Author: jrm5100

docs/release-history.rst

@@ -2,6 +2,11 @@
 Release History
 ===============
 
+v0.9.1 (2021-07-13)
+-------------------
+
+Fix a side-effect of to_plink that swapped allele idxs which could affect edge encoding
+
 v0.9.0 (2021-07-12)
 -------------------
 

pandas_genomics/io/plink/to_plink.py

@@ -164,7 +164,7 @@ def save_bed(data, output_bed):
 
 
 def gt_array_to_plink_bits(gt_series):
-    allele_ids = gt_series.array.allele_idxs
+    allele_ids = gt_series.array.allele_idxs.copy()
     # Replace missing with 0,1
     missing = gt_series.array.is_missing
     allele_ids[missing] = (0, 1)

pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "pandas-genomics"
-version = "0.9.0"
+version = "0.9.1"
 description = "Pandas ExtensionDtypes and ExtensionArray for working with genomics data"
 license = "BSD-3-Clause"
 authors = ["John McGuigan <[email protected]>"]

tests/io/test_plink.py

@@ -1,6 +1,7 @@
 from pathlib import Path
 
 import pytest
+from numpy.testing import assert_array_equal
 from pandas._testing import assert_frame_equal
 
 from pandas_genomics import io, sim
@@ -41,6 +42,7 @@ def test_round_trip_sim(tmp_path):
     d.mkdir()
     output = str(d / "test")
     data = sim.BAMS().generate_case_control()
+    original = data.copy()
     io.to_plink(
         data,
         output,
@@ -59,6 +61,11 @@ def test_round_trip_sim(tmp_path):
         data, loaded_data, check_categorical=False
     )  # Categorical order may be different
 
+    # Ensure there were no side effects
+    assert_array_equal(
+        original["SNP1"].array.allele_idxs, data["SNP1"].array.allele_idxs
+    )
+
 
 @pytest.mark.slow
 def test_loaded_medium():