def test_non_zero_df(self):
tested_object = GeometricMeanNormalization(self.dummy_df,
zero_threshold=80)
data = [[255, 26, 48, 75], [955, 222, 46, 65], [89, 54, 145, 29]]
column_names = ['Sample_1', 'Sample_2', 'Sample_3', 'Sample_4']
ind = ["Gen_1", "Gen_3", 'Gen_4']
expected_result = pd.DataFrame(data, columns=column_names,
index=ind).astype('float')
pd.testing.assert_frame_equal(tested_object.non_zero_df(self.dummy_df),
expected_result)
def test_scaling_factor_zero_thresh_70(self):
tested_object = GeometricMeanNormalization(self.dummy_df,
zero_threshold=70)
data = [3.495248, 0.612726, 0.699505, 0.432524]
ind = ['Sample_1', 'Sample_2', 'Sample_3', 'Sample_4']
expected_result = pd.Series(data, index=ind)
pd.testing.assert_series_equal(tested_object.scaling_factors,
expected_result)
def test_scaling_factor_zero_thresh_80(self):
tested_object = GeometricMeanNormalization(self.dummy_df,
zero_threshold=80)
data = [3.648263, 0.805390, 0.686732, 0.432524]
ind = ['Sample_1', 'Sample_2', 'Sample_3', 'Sample_4']
expected_result = pd.Series(data, index=ind)
pd.testing.assert_series_equal(tested_object.scaling_factors,
expected_result)
def test_removed_zero_df_None(self):
data = [[255, 26, 48, 75], [366, 0, 78, 0], [955, 0, 46, 65],
[89, 54, 145, 29]]
column_names = ['Sample_1', 'Sample_2', 'Sample_3', 'Sample_4']
ind = ["Gen_1", 'Gen_2', "Gen_3", 'Gen_4']
dummy_df = pd.DataFrame(data, columns=column_names, index=ind)
tested_object = GeometricMeanNormalization(dummy_df)
expected_result = None
self.assertEqual(tested_object.removed_zero_df, expected_result)
def test_log_df(self):
input_data = [[255, 26, 48, 75], [955, 222, 46, 65], [89, 54, 145, 29]]
input_column_names = ['Sample_1', 'Sample_2', 'Sample_3', 'Sample_4']
input_ind = ["Gen_1", "Gen_3", 'Gen_4']
input_df = pd.DataFrame(input_data,
columns=input_column_names,
index=input_ind)
tested_object = GeometricMeanNormalization(self.dummy_df)
data = [[5.541264, 3.258097, 3.871201, 4.317488],
[6.861711, 5.402677, 3.828641, 4.174387],
[4.488636, 3.988984, 4.976734, 3.367296]]
column_names = ['Sample_1', 'Sample_2', 'Sample_3', 'Sample_4']
ind = ["Gen_1", "Gen_3", 'Gen_4']
expected_result = pd.DataFrame(data, columns=column_names, index=ind)
pd.testing.assert_frame_equal(tested_object.log_df(input_df),
expected_result)
def test_log_base_n_df(self):
input_data = [[255, 26, 48, 75], [955, 222, 46, 65], [89, 54, 145, 29]]
input_column_names = ['Sample_1', 'Sample_2', 'Sample_3', 'Sample_4']
input_ind = ["Gen_1", "Gen_3", 'Gen_4']
input_df = pd.DataFrame(input_data,
columns=input_column_names,
index=input_ind)
tested_object = GeometricMeanNormalization(self.dummy_df,
log_number=10)
data = [[2.406540, 1.414973, 1.681241, 1.875061],
[2.980003, 2.346353, 1.662758, 1.812913],
[1.949390, 1.732394, 2.161368, 1.462398]]
column_names = ['Sample_1', 'Sample_2', 'Sample_3', 'Sample_4']
ind = ["Gen_1", "Gen_3", 'Gen_4']
expected_result = pd.DataFrame(data, columns=column_names, index=ind)
pd.testing.assert_frame_equal(tested_object.log_df(input_df),
expected_result)
def test_calculating_and_substracting_mean_row(self):
input_data = [[5.541264, 3.258097, 3.871201, 4.317488],
[6.861711, 5.402677, 3.828641, 4.174387],
[4.488636, 3.988984, 4.976734, 3.367296]]
input_column_names = ['Sample_1', 'Sample_2', 'Sample_3', 'Sample_4']
input_ind = ["Gen_1", "Gen_3", 'Gen_4']
input_df = pd.DataFrame(input_data,
columns=input_column_names,
index=input_ind)
tested_object = GeometricMeanNormalization(self.dummy_df)
data = [[1.294251, -0.988916, -0.375811, 0.070475],
[1.794857, 0.335823, -1.238213, -0.892467],
[0.283224, -0.216428, 0.771321, -0.838117]]
column_names = ['Sample_1', 'Sample_2', 'Sample_3', 'Sample_4']
ind = ["Gen_1", "Gen_3", 'Gen_4']
expected_result = pd.DataFrame(data, columns=column_names, index=ind)
pd.testing.assert_frame_equal(
tested_object.calculating_and_substracting_mean_row(
input_df).round(6), expected_result)
def test_normalized_df(self):
tested_object = GeometricMeanNormalization(self.dummy_df,
zero_threshold=100)
data = [[69.896271, 32.282490, 69.896271, 173.400644],
[100.321707, 57.115175, 113.581441, 0.000000],
[261.768388, 275.642802, 66.983926, 150.280558],
[24.395169, 67.048249, 211.144986, 67.048249]]
column_names = ['Sample_1', 'Sample_2', 'Sample_3', 'Sample_4']
ind = ["Gen_1", "Gen_2", "Gen_3", 'Gen_4']
expected_result = pd.DataFrame(data, columns=column_names, index=ind)
pd.testing.assert_frame_equal(tested_object.normalized_df,
expected_result)
def test_normalized_df_replace_0_to_1(self):
tested_object = GeometricMeanNormalization(self.dummy_df,
zero_threshold=100,
replace_0_to_1=True)
data = [[52.757471, 24.026807, 33.691852, 178.111683],
[75.722488, 42.508965, 54.749259, 0.000000],
[197.581903, 205.151963, 32.288025, 154.363458],
[18.413392, 49.901829, 101.777469, 68.869851]]
column_names = ['Sample_1', 'Sample_2', 'Sample_3', 'Sample_4']
ind = ["Gen_1", "Gen_2", "Gen_3", 'Gen_4']
expected_result = pd.DataFrame(data, columns=column_names, index=ind)
pd.testing.assert_frame_equal(tested_object.normalized_df,
expected_result)
def test_scaling_factor_zero_thresh_80_more_zeros(self):
data = [[255, 26, 48, 75], [366, 0, 78, 0], [955, 0, 46, 65],
[89, 54, 145, 29]]
column_names = ['Sample_1', 'Sample_2', 'Sample_3', 'Sample_4']
ind = ["Gen_1", 'Gen_2', "Gen_3", 'Gen_4']
more_zero_example = pd.DataFrame(data, columns=column_names, index=ind)
tested_object = GeometricMeanNormalization(more_zero_example,
zero_threshold=80)
data = [2.487833, 0.588685, 1.424677, 0.752771]
ind = ['Sample_1', 'Sample_2', 'Sample_3', 'Sample_4']
expected_result = pd.Series(data, index=ind)
pd.testing.assert_series_equal(tested_object.scaling_factors,
expected_result)
def test_removed_zero_df(self):
data = [[255, 26, 48, 75], [366, 0, 78, 0], [955, 0, 46, 65],
[89, 54, 145, 29]]
column_names = ['Sample_1', 'Sample_2', 'Sample_3', 'Sample_4']
ind = ["Gen_1", 'Gen_2', "Gen_3", 'Gen_4']
dummy_df = pd.DataFrame(data, columns=column_names, index=ind)
tested_object = GeometricMeanNormalization(dummy_df)
data = [
[366, 0, 78, 0],
[955, 0, 46, 65],
]
column_names = ['Sample_1', 'Sample_2', 'Sample_3', 'Sample_4']
ind = ['Gen_2', "Gen_3"]
expected_result = pd.DataFrame(data, columns=column_names, index=ind)
scaling_factors = tested_object.scaling_factors # noqa
pd.testing.assert_frame_equal(tested_object.removed_zero_df,
expected_result)
def test_multiindex_normalization_second_element(self):
tested_object = pd.DataFrame.from_dict(
{
('genus_1', 'specie_1'): {
'1': 3,
'2': 2,
'3': 1,
'4': 0
},
('genus_1', 'specie_3'): {
'1': 5,
'2': 8,
'3': 4,
'4': 7
},
('genus_2', 'specie_2'): {
'1': 25,
'2': 6,
'3': 3,
'4': 9
},
('genus_2', 'specie_4'): {
'1': 9,
'2': 18,
'3': 6,
'4': 3
},
('genus_3', 'specie_5'): {
'1': 5,
'2': 0,
'3': 3,
'4': 2
},
('genus_3', 'specie_6'): {
'1': 19,
'2': 8,
'3': 0,
'4': 0
},
},
orient='index')
tested_object.columns.name = 'sample'
tested_object.index.set_names(['genus', 'specie'], inplace=True)
normalized_tested_object = GeometricMeanNormalization(
tested_object, normalization_level=1)
expected_result = pd.DataFrame.from_dict(
{
('genus_1', 'specie_1'): {
'1': 2.449490,
'2': 1.446254,
'3': 1.446254,
'4': 0.000000
},
('genus_1', 'specie_3'): {
'1': 4.082483,
'2': 5.785015,
'3': 5.785015,
'4': 6.204679
},
('genus_2', 'specie_2'): {
'1': 20.412415,
'2': 4.338761,
'3': 4.338761,
'4': 7.977444
},
('genus_2', 'specie_4'): {
'1': 7.348469,
'2': 13.016284,
'3': 8.677523,
'4': 2.659148
},
('genus_3', 'specie_5'): {
'1': 4.082483,
'2': 0.000000,
'3': 4.338761,
'4': 1.772765
},
('genus_3', 'specie_6'): {
'1': 15.513435,
'2': 5.785015,
'3': 0.000000,
'4': 0.000000
},
},
orient='index')
expected_result.columns.name = 'sample'
expected_result.index.set_names(['genus', 'specie'], inplace=True)
pd.testing.assert_frame_equal(normalized_tested_object.normalized_df,
expected_result)
def test_multiindex_normalization_first_element(self):
tested_object = pd.DataFrame.from_dict(
{
('genus_1', 'specie_1'): {
'1': 3,
'2': 2,
'3': 1,
'4': 0
},
('genus_1', 'specie_3'): {
'1': 5,
'2': 8,
'3': 4,
'4': 7
},
('genus_2', 'specie_2'): {
'1': 25,
'2': 6,
'3': 3,
'4': 9
},
('genus_2', 'specie_4'): {
'1': 9,
'2': 18,
'3': 6,
'4': 3
},
('genus_3', 'specie_5'): {
'1': 5,
'2': 0,
'3': 3,
'4': 2
},
('genus_3', 'specie_6'): {
'1': 19,
'2': 8,
'3': 0,
'4': 0
},
},
orient='index')
tested_object.columns.name = 'sample'
tested_object.index.set_names(['genus', 'specie'], inplace=True)
normalized_tested_object = GeometricMeanNormalization(
tested_object, normalization_level='genus')
expected_result = pd.DataFrame.from_dict(
{
('genus_1', 'specie_1'): {
'1': 1.520270,
'2': 1.454854,
'3': 1.914414,
'4': 0.000000
},
('genus_1', 'specie_3'): {
'1': 2.533784,
'2': 5.819417,
'3': 7.657658,
'4': 10.050676
},
('genus_2', 'specie_2'): {
'1': 12.668919,
'2': 4.364563,
'3': 5.743243,
'4': 12.922298
},
('genus_2', 'specie_4'): {
'1': 4.560811,
'2': 13.093689,
'3': 11.486487,
'4': 4.307433
},
('genus_3', 'specie_5'): {
'1': 2.533784,
'2': 0.000000,
'3': 5.7432433,
'4': 2.871622
},
('genus_3', 'specie_6'): {
'1': 9.628379,
'2': 5.819417,
'3': 0.000000,
'4': 0.000000
},
},
orient='index')
expected_result.columns.name = 'sample'
expected_result.index.set_names(['genus', 'specie'], inplace=True)
pd.testing.assert_frame_equal(normalized_tested_object.normalized_df,
expected_result)
def test_check_format(self):
tested_object = GeometricMeanNormalization(self.dummy_df)
pd.testing.assert_frame_equal(tested_object.df, self.dummy_df)
