def test_complex_sum_class(self):
out_df = aggregate_at_taxon_level(self.df, taxon_level='class',
simple_aggregation=False,
grouping_cols=self.grouping_cols)
if generate_output_files:
out_df.to_csv('complex_sum_class.csv', index=False)
exp_df = pd.read_csv(self.complex_sum_class)
self.assertTrue(out_df.equals(exp_df))
def test_simple_sum(self):
out_df = aggregate_at_taxon_level(self.df, taxon_level='genus',
simple_aggregation=True,
grouping_cols=self.grouping_cols)
if generate_output_files:
out_df.to_csv('simple_sum.csv', index=False)
exp_df = pd.read_csv(self.simple_sum)
self.assertTrue(out_df.equals(exp_df))
def test_complex_keep_all_cols(self):
out_df = aggregate_at_taxon_level(self.df, taxon_level='genus',
simple_aggregation=False,
grouping_cols=self.grouping_cols,
keep_all_cols=True)
out_columns = set(out_df.columns)
exp_columns = set(self.df.columns)
exp_columns.remove('species')
self.assertEqual(out_columns, exp_columns)
def test_simple_keep_all_cols(self):
out_df = aggregate_at_taxon_level(self.df, taxon_level='genus',
simple_aggregation=True,
grouping_cols=self.grouping_cols,
keep_all_cols=True)
out_columns = set(out_df.columns)
exp_columns = set(self.df.columns).difference(self.taxon_cols)
exp_columns.add('genus')
self.assertEqual(out_columns, exp_columns)
def test_custom_dummy_values(self):
temp_df = self.df.copy()
temp_df.loc[temp_df['genus'] == 'g__', 'genus'] = 'CUSTOM'
print(temp_df)
out_df = aggregate_at_taxon_level(temp_df, taxon_level='genus',
dummy_values={'genus': 'CUSTOM'},
simple_aggregation=True,
grouping_cols=self.grouping_cols)
if generate_output_files:
out_df.to_csv('simple_sum_custom_dummy.csv', index=False)
exp_df = pd.read_csv(self.simple_sum_custom_dummy)
self.assertTrue(out_df.equals(exp_df))
def test_not_all_required_grouping_cols_in_df(self):
self.df.rename(columns={'experiment_id': 'NONE'}, inplace=True)
with self.assertRaises(ValueError):
aggregate_at_taxon_level(self.df, taxon_level='genus',
simple_aggregation=True,
grouping_cols=self.grouping_cols)
def test_no_taxon_level_in_df(self):
self.df.rename(columns={'genus': 'no_genus'}, inplace=True)
with self.assertRaises(ValueError):
aggregate_at_taxon_level(self.df, taxon_level='genus',
simple_aggregation=True,
grouping_cols=self.grouping_cols)
def test_unrecognized_taxon_level(self):
with self.assertRaises(ValueError):
aggregate_at_taxon_level(self.df, taxon_level='gibberish',
simple_aggregation=True,
grouping_cols=self.grouping_cols)
def test_unrecognized_dummy_value_key(self):
with self.assertRaises(ValueError):
aggregate_at_taxon_level(self.df, taxon_level='genus',
dummy_values={'gibberish': 'g__'},
simple_aggregation=True,
grouping_cols=self.grouping_cols)