def produce(self, unknown_args=None):
opts = self.namespace
name = opts.name or self.name
if opts.files:
# load all files
sample_collections = []
if callable(opts.header):
opts.header = [opts.header(f) for f in opts.files]
for i, file_obj in enumerate(opts.files):
use_header = isinstance(opts.header[i], int)
sample_collections.append(
SampleCollection.from_file(
f'Sample collection, part {i} of {name}',
file_obj,
columns_selector=opts.columns[i].get_iterator
if opts.columns else None,
samples=opts.samples[i] if opts.samples else None,
reverse_selection=getattr(opts, 'reverse', False),
delimiter=opts.delimiter,
header_line=opts.header[i] if use_header else None,
use_header=use_header,
prefix=opts.header[i] if not use_header else None,
description_column=opts.description_column))
opts.sample_collection = sum(sample_collections,
SampleCollection(name))
return opts
def minimal_data():
tp53 = Gene('TP53')
map2k1 = Gene('MAP2K1')
case = SampleCollection('case', [Sample('1', {tp53: 2, map2k1: 1})])
control = SampleCollection('control', [Sample('1', {tp53: 1, map2k1: 1})])
return tp53, map2k1, case, control
def test_ttest():
data1 = {'BAD': 1.2345, 'FUCA2': 6.5432}
data2 = {'BAD': 2.3456, 'FUCA2': 7.6543}
data3 = {'BAD': 6.3456, 'FUCA2': 11.6543}
data4 = {'BAD': 7.1111, 'FUCA2': 9.9711}
tumour_samples = [Sample.from_names('Tumour_1', data1), Sample.from_names('Tumour_2', data2)]
normal_samples = [Sample.from_names('Normal_1', data3), Sample.from_names('Normal_2', data4)]
tumour = SampleCollection('Tumour', tumour_samples)
normal = SampleCollection('Normal', normal_samples)
experiment = Experiment(case=tumour, control=normal)
tt = ttest(experiment)
assert isinstance(tt, pd.Series)
assert all(gene in list(tt.keys()) for gene in experiment.get_all().genes)
def test_from_csv():
with temp_text_file(csv_contents) as csv_file:
collection = SampleCollection.from_csv_file('all_samples.csv',
csv_file)
assert len(collection.samples) == 4
with temp_text_file(csv_contents) as csv_file:
with warns(
UserWarning,
match=
'You are using not comma delimiter for what looks like csv file.'
):
SampleCollection.from_csv_file('all_samples.csv',
csv_file,
delimiter='\t')
def test_init():
genes1 = {Gene('BAD'): 1.2345, Gene('FUCA2'): 6.5432}
genes2 = {Gene('BAD'): 2.3456, Gene('FUCA2'): 7.6543}
samples = [Sample('Tumour_1', genes1), Sample('Tumour_2', genes2)]
sample_collection = SampleCollection('Tumour', samples)
assert sample_collection.name == 'Tumour'
assert all(isinstance(k, Sample) for k in sample_collection.samples)
def test_get_all():
data1 = {'BAD': 1.2345, 'FUCA2': 6.5432}
data2 = {'BAD': 2.3456, 'FUCA2': 7.6543}
data3 = {'BAD': 3.4567}
tumour_samples = [
Sample.from_names('Tumour_1', data1),
Sample.from_names('Tumour_2', data2)
]
normal_samples = [Sample.from_names('Normal_1', data3)]
tumour = SampleCollection('Tumour', tumour_samples)
normal = SampleCollection('Normal', normal_samples)
experiment_samples = Experiment(case=tumour, control=normal).get_all()
assert isinstance(experiment_samples, SampleCollection)
assert all(label in experiment_samples.labels
for label in tumour.labels + normal.labels)
def test_init():
data1 = {'BAD': 1.2345, 'FUCA2': 6.5432}
data2 = {'BAD': 2.3456, 'FUCA2': 7.6543}
data3 = {'BAD': 3.4567}
tumour_samples = [
Sample.from_names('Tumour_1', data1),
Sample.from_names('Tumour_2', data2)
]
normal_samples = [Sample.from_names('Normal_1', data3)]
tumour = SampleCollection('Tumour', tumour_samples)
normal = SampleCollection('Normal', normal_samples)
experiment = Experiment(case=tumour, control=normal)
assert isinstance(experiment.case, SampleCollection)
assert isinstance(experiment.control, SampleCollection)
assert experiment.case == tumour
assert experiment.control == normal
def test_from_gct():
with temp_text_file(gct_contents) as gct_file:
expected_warning = 'Samples count \(4\) does not match with the 3 declared in all_samples.gct file.'
with warns(UserWarning, match=expected_warning):
collection = SampleCollection.from_gct_file(
'all_samples.gct', gct_file)
assert len(collection.samples) == 4
assert collection.labels == ['NORM-1', 'GBM-1', 'GBM-2', 'OV-1']
# replace version definition
lines = gct_contents.split('\n')
lines[0] = '#1.1'
old_content = '\n'.join(lines)
with temp_text_file(old_content) as old_gct_file:
with warns(UserWarning, match='Unsupported version of GCT file'):
SampleCollection.from_gct_file('Outdated file', old_gct_file)
def shuffle_and_divide(merged_collection, midpoint):
shuffled = copy(merged_collection.samples)
shuffle(shuffled)
return (SampleCollection('Random collection', shuffled[:midpoint]),
SampleCollection('Random collection', shuffled[midpoint:]))