Add tests for loading files and fix an issue with default alleles

pandas_genomics/io/plink.py

@@ -70,16 +70,18 @@ def from_plink(bed_file: str):
         for v_idx in range(num_variants):
             variant_info_dict = variant_info.iloc[v_idx].to_dict()
             variant_id = str(variant_info_dict['variant_id'])
+            a1 = str(variant_info_dict['allele1'])
+            a2 = str(variant_info_dict['allele2'])
             variant = Variant(variant_id=variant_id,
                               chromosome=str(variant_info_dict['chromosome']),
                               coordinate=int(variant_info_dict['coordinate']),
-                              alleles=[str(variant_info_dict['allele1']), str(variant_info_dict['allele2'])])
+                              alleles=[a1, a2])
             genotypes = []
             chunk = f.read(chunk_size)  # Encoded chunk of results for each variant
             for byte in chunk:
                 # for each byte, get 2 bits at a time in reverse order (as a string, so '00', '01', '10', or '11')
                 bitstrings = [f"{byte:08b}"[i:i+2] for i in range(0, 8, 2)][::-1]
-                genotypes.extend([variant.make_genotype_from_plink_bits(bs) for bs in bitstrings])
+                genotypes.extend([variant.make_genotype_from_plink_bits(bs, a1, a2) for bs in bitstrings])
             # Remove nonexistent samples at the end
             genotypes = genotypes[:num_samples]
             df[variant_id] = GenotypeArray(values=genotypes, dtype=GenotypeDtype(variant))

pandas_genomics/scalars.py

@@ -225,14 +225,21 @@ def make_genotype_from_str(self, gt_str: str, sep: str = "/") -> 'Genotype':
         a2 = self.get_allele_idx(allele2, add=True)
         return Genotype(self, a1, a2)
 
-    def make_genotype_from_plink_bits(self, plink_bits: str) -> 'Genotype':
+    def make_genotype_from_plink_bits(self,
+                                      plink_bits: str,
+                                      allele1: str,
+                                      allele2: str) -> 'Genotype':
         """
         Create a genotype from PLINK Bed file bits
 
         Parameters
         ----------
         plink_bits: str
             A string with allele indices as encoded in plink format, one of {'00', '01', '10', '11'}
+        allele1: str
+            Allele corresponding to the first allele in the plink file
+        allele2: str
+            Allele corresponding to the second allele in the plink file
 
         Returns
         -------

tests/conftest.py

@@ -1,59 +1,21 @@
 from pandas.tests.extension.conftest import *
 
 
-# Below fixtures are copied from pandas.conftest
-# They could be imported, but that would require having hypothesis as a dependency
-@pytest.fixture(params=[None, lambda x: x])
-def sort_by_key(request):
-    """
-    Simple fixture for testing keys in sorting methods.
-    Tests None (no key) and the identity key.
-    """
-    return request.param
+def pytest_addoption(parser):
+    parser.addoption(
+        "--runslow", action="store_true", default=False, help="run slow tests"
+    )
 
 
-@pytest.fixture(params=["__eq__", "__ne__", "__le__", "__lt__", "__ge__", "__gt__"])
-def all_compare_operators(request):
-    """
-    Fixture for dunder names for common compare operations
-    * >=
-    * >
-    * ==
-    * !=
-    * <
-    * <=
-    """
-    return request.param
+def pytest_configure(config):
+    config.addinivalue_line("markers", "slow: mark test as slow to run")
 
 
-_all_numeric_reductions = [
-    "sum",
-    "max",
-    "min",
-    "mean",
-    "prod",
-    "std",
-    "var",
-    "median",
-    "kurt",
-    "skew",
-]
-
-
-@pytest.fixture(params=_all_numeric_reductions)
-def all_numeric_reductions(request):
-    """
-    Fixture for numeric reduction names.
-    """
-    return request.param
-
-
-_all_boolean_reductions = ["all", "any"]
-
-
-@pytest.fixture(params=_all_boolean_reductions)
-def all_boolean_reductions(request):
-    """
-    Fixture for boolean reduction names.
-    """
-    return request.param
+def pytest_collection_modifyitems(config, items):
+    if config.getoption("--runslow"):
+        # --runslow given in cli: do not skip slow tests
+        return
+    skip_slow = pytest.mark.skip(reason="need --runslow option to run")
+    for item in items:
+        if "slow" in item.keywords:
+            item.add_marker(skip_slow)

tests/genotype_array/conftest.py

@@ -0,0 +1,59 @@
+import pytest
+
+
+# Below fixtures are copied from pandas.conftest
+# They could be imported, but that would require having hypothesis as a dependency
+@pytest.fixture(params=[None, lambda x: x])
+def sort_by_key(request):
+    """
+    Simple fixture for testing keys in sorting methods.
+    Tests None (no key) and the identity key.
+    """
+    return request.param
+
+
+@pytest.fixture(params=["__eq__", "__ne__", "__le__", "__lt__", "__ge__", "__gt__"])
+def all_compare_operators(request):
+    """
+    Fixture for dunder names for common compare operations
+    * >=
+    * >
+    * ==
+    * !=
+    * <
+    * <=
+    """
+    return request.param
+
+
+_all_numeric_reductions = [
+    "sum",
+    "max",
+    "min",
+    "mean",
+    "prod",
+    "std",
+    "var",
+    "median",
+    "kurt",
+    "skew",
+]
+
+
+@pytest.fixture(params=_all_numeric_reductions)
+def all_numeric_reductions(request):
+    """
+    Fixture for numeric reduction names.
+    """
+    return request.param
+
+
+_all_boolean_reductions = ["all", "any"]
+
+
+@pytest.fixture(params=_all_boolean_reductions)
+def all_boolean_reductions(request):
+    """
+    Fixture for boolean reduction names.
+    """
+    return request.param

tests/io/conftest.py

@@ -0,0 +1,21 @@
+from pathlib import Path
+import pytest
+
+from pandas_genomics import io
+
+data_dir = Path(__file__).parent.parent / "data"
+
+
+@pytest.fixture
+def plink_small():
+    bed_file = data_dir / "plink" / "plink_test_small.bed"
+    result = io.from_plink(bed_file)
+    return result
+
+
+@pytest.mark.slow
+@pytest.fixture
+def plink_medium():
+    bed_file = data_dir / "plink" / "plink_test_medium.bed"
+    result = io.from_plink(bed_file)
+    return result

tests/io/test_io.py

@@ -0,0 +1,14 @@
+import pytest
+
+
+def test_loaded_small(plink_small):
+    """Validate the small dataset"""
+    # TODO: Add more assertions
+    assert plink_small.shape == (150, 3020)
+
+
+@pytest.mark.slow
+def test_loaded_medium(plink_medium):
+    """Validate the medium dataset"""
+    # TODO: Add more assertions
+    assert plink_medium.shape == (600, 45100)