def setUp(self):
# create a couple Observations to use and a corresponding serializer
self.el1 = Observation('sampleA', {'phenotype': StringAttribute('WT')})
self.el1_serializer = ObservationSerializer(self.el1)
self.el2 = Observation('sampleB', {'phenotype': StringAttribute('KO')})
self.el2_serializer = ObservationSerializer(self.el2)
# a duplicate of el1 above, for testing addition of duplicate elements:
self.duplicate_element = Observation('sampleA', {})
self.dup_element_serializer = ObservationSerializer(
self.duplicate_element)
# the correct serialized representation of an ElementSet instance
self.expected_element_set_data = {
'multiple': True,
'elements': [self.el1_serializer.data, self.el2_serializer.data]
}
# a correctly formed instance of an ObservationSet
self.element_set = ObservationSet([self.el1, self.el2])
# the class that will execute the tests
self.tester_class = ElementSetSerializerTester(
ObservationSetSerializer)
def setUp(self):
# create a couple Observations to use
self.el1 = Observation('sampleA', {'phenotype': StringAttribute('WT')})
self.el2 = Observation('sampleB', {'phenotype': StringAttribute('KO')})
# a duplicate of element above:
self.duplicate_element = Observation('sampleA', {})
# instantiate the class that will actually execute the tests
self.tester_class = ElementSetTester(ObservationSet)
def setUp(self):
# create a couple Featires to use
self.el1 = Feature('geneA', {'oncogene': StringAttribute('Y')})
self.el2 = Feature('sampleB', {'oncogene': StringAttribute('N')})
# a duplicate of element above:
self.duplicate_element = Feature('geneA', {})
# instantiate the class that will actually execute the tests
self.tester_class = ElementSetTester(FeatureSet)
def test_string_attribute(self):
# this is sort of double test-coverage, but that can't hurt
s = StringAttribute('abc')
self.assertEqual(s.value, 'abc')
s = StringAttribute('a string with space')
self.assertEqual(s.value, 'a_string_with_space')
with self.assertRaises(AttributeValueError):
StringAttribute('-9abc')
with self.assertRaises(AttributeValueError):
StringAttribute(3.4)
def create_feature_set():
# create a couple Features to use and a corresponding serializer
el1 = Feature('featureA', {'pathway': StringAttribute('foo')})
el1_serializer = FeatureSerializer(el1)
el2 = Feature('sampleB', {'pathway': StringAttribute('bar')})
el2_serializer = FeatureSerializer(el2)
# the correct serialized representation of an ElementSet instance
feature_set_data = {
'multiple': True,
'elements': [el1_serializer.data, el2_serializer.data]
}
return feature_set_data
def create_observation_set():
# create a couple Observations to use and a corresponding serializer
el1 = Observation('sampleA', {'phenotype': StringAttribute('WT')})
el1_serializer = ObservationSerializer(el1)
el2 = Observation('sampleB', {'phenotype': StringAttribute('KO')})
el2_serializer = ObservationSerializer(el2)
# the correct serialized representation of an ElementSet instance
observation_set_data = {
'multiple': True,
'elements': [el1_serializer.data, el2_serializer.data]
}
return observation_set_data
def test_metadata_correct(self):
resource_path = os.path.join(TESTDIR, 'gene_annotations.tsv')
t = FeatureTable()
column_dict = {}
feature_list = []
for i, line in enumerate(open(resource_path)):
if i == 0:
contents = line.strip().split('\t')
for j,c in enumerate(contents[1:]):
column_dict[j] = c
else:
contents = line.strip().split('\t')
gene_name = contents[0]
attr_dict = {}
for j,v in enumerate(contents[1:]):
try:
v = int(v)
attr = IntegerAttribute(v)
except ValueError:
attr = StringAttribute(v)
attr_dict[column_dict[j]] = attr
f = Feature(gene_name, attr_dict)
feature_list.append(f)
expected_feature_set = FeatureSetSerializer(FeatureSet(feature_list)).data
metadata = t.extract_metadata(resource_path)
self.assertEqual(metadata[FEATURE_SET_KEY], expected_feature_set)
self.assertIsNone(metadata[OBSERVATION_SET_KEY])
self.assertIsNone(metadata[PARENT_OP_KEY])
def test_metadata_correct(self):
resource_path = os.path.join(TESTDIR, 'gene_annotations.tsv')
t = FeatureTable()
column_dict = {}
feature_list = []
for i, line in enumerate(open(resource_path)):
if i == 0:
contents = line.strip().split('\t')
for j, c in enumerate(contents[1:]):
column_dict[j] = c
else:
contents = line.strip().split('\t')
gene_name = contents[0]
attr_dict = {}
for j, v in enumerate(contents[1:]):
try:
v = int(v)
attr = IntegerAttribute(v)
except ValueError:
attr = StringAttribute(v)
attr_dict[column_dict[j]] = attr
f = Feature(gene_name, attr_dict)
feature_list.append(f)
expected_feature_set = FeatureSetSerializer(
FeatureSet(feature_list)).data
metadata = t.extract_metadata(resource_path, 'tsv')
# Commented out when we removed the automatic creation of Feature metadata
# for FeatureTable resource types. For large files, it was causing issues
# with exceptionally large JSON failing to store in db table.
#self.assertEqual(metadata[FEATURE_SET_KEY], expected_feature_set)
self.assertIsNone(metadata[FEATURE_SET_KEY])
self.assertIsNone(metadata[OBSERVATION_SET_KEY])
self.assertIsNone(metadata[PARENT_OP_KEY])
def test_merge_of_different_types_fails(self):
'''
We cannot merge two different types (e.g. and Obs Set and Feat. Set)
Test that it raises an exception.
'''
element_list1 = [self.el1, self.el2]
some_feature = Feature('geneA', {'oncogene': StringAttribute('Y')})
element_list2 = [
some_feature,
]
obs_set = ObservationSet(element_list1)
feature_set = FeatureSet(element_list2)
with self.assertRaises(Exception):
new_set = merge_element_set([obs_set, feature_set])
def test_adding_new_attribute(self, testcase):
'''
Test adding a new attribute
'''
float_attr = FloatAttribute(0.01)
int_attr = IntegerAttribute(3)
element = self.element_class('some_identifier', {
'keyA': float_attr,
'keyB': int_attr
})
element.add_attribute('keyC', {
'attribute_type': 'String',
'value': 'abc'
})
expected_attr = StringAttribute('abc')
testcase.assertEqual(element.attributes['keyC'], expected_attr)
expected_keys = set(['keyA', 'keyB', 'keyC'])
existing_keys = set(element.attributes.keys())
testcase.assertTrue(expected_keys == existing_keys)
def test_metadata_correct(self):
resource_path = os.path.join(TESTDIR, 'three_column_annotation.tsv')
t = AnnotationTable()
column_dict = {}
obs_list = []
for i, line in enumerate(open(resource_path)):
if i == 0:
contents = line.strip().split('\t')
for j,c in enumerate(contents[1:]):
column_dict[j] = c
else:
contents = line.strip().split('\t')
samplename = contents[0]
attr_dict = {}
for j,v in enumerate(contents[1:]):
attr = StringAttribute(v)
attr_dict[column_dict[j]] = attr
obs = Observation(samplename, attr_dict)
obs_list.append(obs)
expected_obs_set = ObservationSetSerializer(ObservationSet(obs_list)).data
metadata = t.extract_metadata(resource_path)
self.assertEqual(metadata[OBSERVATION_SET_KEY], expected_obs_set)
self.assertIsNone(metadata[FEATURE_SET_KEY])
self.assertIsNone(metadata[PARENT_OP_KEY])
def convert(self, input_key, user_input, op_dir, staging_dir):
s = StringAttribute(user_input)
return {input_key: s.value}
def setUp(self):
self.establish_clients()
self.new_resource1 = Resource.objects.create(
name = 'foo.txt',
owner = self.regular_user_1,
is_active=True
)
self.new_resource2 = Resource.objects.create(
name = 'bar.txt',
owner = self.regular_user_1,
is_active=True
)
self.new_resource3 = Resource.objects.create(
name = 'baz.txt',
owner = self.regular_user_1,
is_active=True
)
# create a workspace to which we will eventually add resources
self.workspace = Workspace.objects.create(
owner = self.regular_user_1
)
self.empty_workspace = Workspace.objects.create(
owner = self.regular_user_1
)
# create a few Observations to use with the different Resources
obs1 = Observation('sampleA', {
'phenotype': StringAttribute('WT')
})
obs1_serializer = ObservationSerializer(obs1)
obs2 = Observation('sampleB', {
'phenotype': StringAttribute('KO')
})
obs2_serializer = ObservationSerializer(obs2)
obs3 = Observation('sampleC', {
'phenotype': StringAttribute('KO')
})
obs3_serializer = ObservationSerializer(obs3)
# create Features to use and a corresponding serializer
feature1 = Feature('featureA', {
'pathway': StringAttribute('foo')
})
feature1_serializer = FeatureSerializer(feature1)
feature2 = Feature('featureB', {
'pathway': StringAttribute('bar')
})
feature2_serializer = FeatureSerializer(feature2)
feature3 = Feature('featureC', {
'pathway': StringAttribute('bar3')
})
feature3_serializer = FeatureSerializer(feature3)
feature4 = Feature('featureD', {
'pathway': StringAttribute('bar')
})
feature4_serializer = FeatureSerializer(feature4)
# create an ObservationSet for resource1
observation_set_data1 = {
'multiple': True,
'elements': [
obs1_serializer.data,
obs2_serializer.data
]
}
# create an ObservationSet for resource2
observation_set_data2 = {
'multiple': True,
'elements': [
obs3_serializer.data,
]
}
# create a FeatureSet for resource1
feature_set_data1 = {
'multiple': True,
'elements': [
feature1_serializer.data,
feature2_serializer.data
]
}
# create a FeatureSet for resource2
feature_set_data2 = {
'multiple': True,
'elements': [
feature3_serializer.data,
feature4_serializer.data
]
}
metadata1 = {
RESOURCE_KEY: self.new_resource1.pk,
OBSERVATION_SET_KEY: observation_set_data1,
FEATURE_SET_KEY: feature_set_data1,
PARENT_OP_KEY: None
}
metadata2 = {
RESOURCE_KEY: self.new_resource2.pk,
OBSERVATION_SET_KEY: observation_set_data2,
FEATURE_SET_KEY: feature_set_data2,
PARENT_OP_KEY: None
}
rms1 = ResourceMetadataSerializer(data=metadata1)
if rms1.is_valid(raise_exception=True):
rms1.save()
rms2 = ResourceMetadataSerializer(data=metadata2)
if rms2.is_valid(raise_exception=True):
rms2.save()