Support non-numeric fields in input data.

Author: cmclean

nonlinear-covariate-gwas/__init__.py

@@ -26,4 +26,4 @@
 # OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 """DeepNull."""
 
-__version__ = '0.1.2'
+__version__ = '0.1.3'

nonlinear-covariate-gwas/data.py

@@ -169,11 +169,17 @@ def write_plink_or_bolt_file(input_df: pd.DataFrame,
     for column in df.columns:
       values = df[column]
       mask = ~values.isnull()
-      if (values[mask] == values[mask].astype(int)).all():
-        # All non-null values are integers. Convert to the 'Int64' type that
-        # allows nullable integers. This requires nulls to use the pd.NA value
-        # rather than np.nan.
-        df[column] = values.fillna(pd.NA).astype('Int64')
+      try:
+        int_values = values[mask].astype(int)
+      except ValueError:
+        # This is a non-numeric field, leave it as-is.
+        continue
+      else:
+        if (values[mask] == int_values).all():
+          # All non-null values are integers. Convert to the 'Int64' type that
+          # allows nullable integers. This requires nulls to use the pd.NA value
+          # rather than np.nan.
+          df[column] = values.fillna(pd.NA).astype('Int64')
 
   return df.to_csv(
       path_or_buf, sep='\t', index=False, na_rep=str(missing_value))

nonlinear-covariate-gwas/data_test.py

@@ -164,7 +164,7 @@ def test_load_invalid_plink_or_bolt_file(self, str_contents, msg):
   def test_write_plink_or_bolt_file(self):
     df = pd.DataFrame(
         {
-            'FID': [1, 2, 3, 4, 5],
+            'FID': ['fam1', 'fam1', 'fam1', 'fam2', 'fam3'],
             'IID': [1, 2, 3, 4, 5],
             'age': [45, 50, 55, 60, 65],
             'sex': [0, 0, 1, 1, 0],
@@ -175,11 +175,11 @@ def test_write_plink_or_bolt_file(self):
     binary_column_map = {'sex': {0: 1, 1: 2}, 'binary_miss': {0.: 1., 1.: 2.}}
 
     expected = ('FID\tIID\tage\tsex\tbinary_miss\tcont_miss\n'
-                '1\t1\t45\t1\t1\t0.5\n'
-                '2\t2\t50\t1\t2\t1.5\n'
-                '3\t3\t55\t2\t2\t2.5\n'
-                '4\t4\t60\t2\tNA\t3.5\n'
-                '5\t5\t65\t1\t1\tNA\n')
+                'fam1\t1\t45\t1\t1\t0.5\n'
+                'fam1\t2\t50\t1\t2\t1.5\n'
+                'fam1\t3\t55\t2\t2\t2.5\n'
+                'fam2\t4\t60\t2\tNA\t3.5\n'
+                'fam3\t5\t65\t1\t1\tNA\n')
 
     actual = data.write_plink_or_bolt_file(
         df,

nonlinear-covariate-gwas/setup.py

@@ -17,7 +17,7 @@
     #
     # $ pip install deepnull
     name='deepnull',
-    version='0.1.2',  # Keep in sync with __init__.__version__.
+    version='0.1.3',  # Keep in sync with __init__.__version__.
     description='Models nonlinear interactions between covariates and phenotypes',
     long_description=long_description,
     long_description_content_type='text/markdown',

nonlinear-covariate-gwas/train_eval_test.py

@@ -194,6 +194,7 @@ def test_create_deepnull_prediction(self, target_is_binary):
     design_df = _create_df(target_is_binary=target_is_binary, size=size)
     design_df['FID'] = np.arange(size)
     design_df['IID'] = np.arange(size)
+    design_df['unused_str_column'] = np.random.choice(list('abcdefg'), size)
 
     input_df = design_df.copy(deep=True)
     with tempfile.TemporaryDirectory() as tmpdir: