def test_create_nested_json(self):
"""Test that we can create nested json."""
property_tree = GenomePropertiesTree(*self.properties)
json_data = property_tree.create_nested_json(as_dict=True)
root_id = json_data['id']
"""Root could be either GenProp0002 or GenProp0003. See Note 1 in test_find_root_node()."""
self.assertIn(root_id, ['GenProp0002', 'GenProp0003'])
tree_level_one_children = json_data['children']
self.assertEqual(len(tree_level_one_children), 1)
level_one_child = tree_level_one_children[0]
self.assertEqual(level_one_child['id'], 'GenProp0066')
tree_level_two_children = level_one_child['children']
self.assertEqual(len(tree_level_two_children), 2)
level_two_child_one = tree_level_two_children[0]
level_two_child_two = tree_level_two_children[1]
self.assertIn(level_two_child_one['id'],
['GenProp0089', 'GenProp0092'])
self.assertIn(level_two_child_two['id'],
['GenProp0089', 'GenProp0092'])
self.assertNotEqual(level_two_child_one['id'],
level_two_child_two['id'])
self.assertEqual(level_two_child_one['children'], [])
self.assertEqual(level_two_child_two['children'], [])
def test_json_string_creation_nodes_and_links(self):
property_tree = GenomePropertiesTree(*self.properties)
test_json = property_tree.to_json(nodes_and_links=True)
expected_json = '''{
"nodes": [{"id": "GenProp0002", "name": "Coenzyme F420 utilization", "type": "GUILD", "description": null,
"notes": null},
{"id": "GenProp0003", "name": "Coenzyme F420 utilization", "type": "GUILD", "description": null,
"notes": null},
{"id": "GenProp0066", "name": "Coenzyme F420 utilization", "type": "GUILD", "description": null,
"notes": null},
{"id": "GenProp0089", "name": "Coenzyme F420 utilization", "type": "GUILD", "description": null,
"notes": null},
{"id": "GenProp0092", "name": "Coenzyme F420 utilization", "type": "GUILD", "description": null,
"notes": null}],
"links": [{"parent": "GenProp0002", "child": "GenProp0066"}, {"parent": "GenProp0003", "child": "GenProp0066"},
{"parent": "GenProp0066", "child": "GenProp0089"},
{"parent": "GenProp0066", "child": "GenProp0092"}]}'''
test_json_parsed = json.loads(test_json)
expected_json_parsed = json.loads(expected_json)
test_json_nodes = test_json_parsed['nodes']
expected_json_nodes = expected_json_parsed['nodes']
test_json_links = test_json_parsed['links']
expected_json_links = expected_json_parsed['links']
self.assertCountEqual(test_json_nodes, expected_json_nodes)
self.assertCountEqual(test_json_links, expected_json_links)
def setUpClass(cls):
"""Set up testing data for testing."""
"""
Test Properties Rooted DAG Structure:
--> GenProp0089
GenProp0066
--> GenProp0092
"""
property_rows_one = [('AC', 'GenProp0066'),
('DE', 'Coenzyme F420 utilization'),
('TP', 'GUILD'), ('--', ''), ('SN', '1'),
('ID', 'Selfish genetic elements'), ('RQ', '0'),
('EV', 'GenProp0089;')]
property_rows_two = [('AC', 'GenProp0089'),
('DE', 'Coenzyme F420 utilization'),
('TP', 'GUILD'), ('--', ''), ('SN', '1'),
('ID', 'LLM-family F420-associated subfamilies'),
('RQ', '0'),
('EV', 'IPR019910; TIGR03564; sufficient;')]
property_one = parse_genome_property(property_rows_one)
property_two = parse_genome_property(property_rows_two)
cls.tree = GenomePropertiesTree(property_one, property_two)
def test_build_genome_property_connections(self):
"""Test that we can add child and parent genome properties."""
test_properties = self.properties
property_one = test_properties[0]
property_two = test_properties[1]
property_three = test_properties[2]
property_four = test_properties[3]
property_five = test_properties[4]
property_tree = GenomePropertiesTree(*test_properties)
self.assertEqual(property_tree[property_one.id].children[0],
property_three)
self.assertEqual(property_tree[property_two.id].children[0],
property_three)
self.assertCountEqual(property_tree[property_three.id].parents,
[property_one, property_two])
self.assertCountEqual(property_tree[property_three.id].children,
[property_four, property_five])
self.assertEqual(property_tree[property_four.id].parents[0],
property_three)
self.assertEqual(property_tree[property_five.id].parents[0],
property_three)
def test_find_leaf_nodes(self):
"""Test we can find the right leaf nodes."""
property_tree = GenomePropertiesTree(*self.properties)
leaf_ids = [leaf.id for leaf in property_tree.leafs]
self.assertCountEqual(leaf_ids, ['GenProp0089', 'GenProp0092'])
def test_get_property_identifiers(self):
"""Test that we can get a set of all property identifiers in the tree."""
property_tree = GenomePropertiesTree(*self.properties)
self.assertEqual(
property_tree.genome_property_identifiers, {
'GenProp0002', 'GenProp0089', 'GenProp0066', 'GenProp0003',
'GenProp0092'
})
def test_find_root_node(self):
"""Test that we can find the correct genome property root."""
property_tree = GenomePropertiesTree(*self.properties)
root = property_tree.root
"""
Note 2: The root could be GenProp0002 or GenProp0003 depending on which one is stored first in the property
tree. Since we are using a dict not an OrderedDict inside of GenomePropertyTree we cannot guarantee that
GenProp0002 will always be returned as root. Thus we check if the root node is either property.
"""
self.assertIn(root.id, ['GenProp0002', 'GenProp0003'])
def test_json_string_creation(self):
property_tree = GenomePropertiesTree(*self.properties)
test_json = property_tree.to_json()
expected_json_one = '''{"id": "GenProp0002", "name": "Coenzyme F420 utilization", "type": "GUILD", "description": null, "notes": null,
"children": [{"id": "GenProp0066", "name": "Coenzyme F420 utilization", "type": "GUILD", "description": null,
"notes": null, "children": [
{"id": "GenProp0089", "name": "Coenzyme F420 utilization", "type": "GUILD", "description": null,
"notes": null, "children": []},
{"id": "GenProp0092", "name": "Coenzyme F420 utilization", "type": "GUILD", "description": null,
"notes": null, "children": []}]}]}'''
test_json_parsed = json.loads(test_json)
expected_json_parsed_one = json.loads(expected_json_one)
"""Root could be either GenProp0002 or GenProp0003. See Note 1 in test_find_root_node()."""
expected_json_parsed_two = deepcopy(expected_json_parsed_one)
expected_json_parsed_two['id'] = 'GenProp0003'
self.assertIn(test_json_parsed,
[expected_json_parsed_one, expected_json_parsed_two])
def test_create_json_graph_nodes(self):
"""Test that we can create nodes json."""
property_tree = GenomePropertiesTree(*self.properties)
json_nodes = property_tree.create_graph_nodes_json(as_list=True)
ids = {node['id'] for node in json_nodes}
names = {node['name'] for node in json_nodes}
types = {node['type'] for node in json_nodes}
descriptions = {node['description'] for node in json_nodes}
notes = {node['notes'] for node in json_nodes}
self.assertCountEqual(
ids, {
'GenProp0002', 'GenProp0003', 'GenProp0066', 'GenProp0089',
'GenProp0092'
})
self.assertEqual(names, {'Coenzyme F420 utilization'})
self.assertEqual(types, {'GUILD'})
self.assertEqual(descriptions, {None})
self.assertEqual(notes, {None})
def test_create_json_graph_links(self):
"""Test that we can create parent child link json."""
property_tree = GenomePropertiesTree(*self.properties)
json_links = property_tree.create_graph_links_json(as_list=True)
predicted_links = [{
'parent': 'GenProp0002',
'child': 'GenProp0066'
}, {
'parent': 'GenProp0003',
'child': 'GenProp0066'
}, {
'parent': 'GenProp0066',
'child': 'GenProp0089'
}, {
'parent': 'GenProp0066',
'child': 'GenProp0092'
}]
self.assertCountEqual(json_links, predicted_links)
def setUpClass(cls):
"""Set up testing data for testing."""
prebuilt_cache = AssignmentCache()
prebuilt_cache.cache_property_assignment('GenProp0053', 'YES')
prebuilt_cache.cache_property_assignment('GenProp0052', 'NO')
prebuilt_cache.cache_property_assignment('GenProp0051', 'PARTIAL')
prebuilt_cache.cache_step_assignment('GenProp0053', 1, 'YES')
prebuilt_cache.cache_step_assignment('GenProp0053', 2, 'NO')
prebuilt_cache.cache_step_assignment('GenProp0053', 3, 'YES')
cls.cache = prebuilt_cache
"""
Test Properties Rooted DAG Structure:
--> GenProp0089
GenProp0066
--> GenProp0092
"""
property_rows_one = [('AC', 'GenProp0066'),
('DE', 'Coenzyme F420 utilization'),
('TP', 'GUILD'), ('--', ''), ('SN', '1'),
('ID', 'Selfish genetic elements'), ('RQ', '0'),
('EV', 'GenProp0089;'), ('--', ''), ('SN', '2'),
('ID', 'Selfish genetic elements'), ('RQ', '0'),
('EV', 'GenProp0092;')]
property_rows_two = [('AC', 'GenProp0089'),
('DE', 'Coenzyme F420 utilization'),
('TP', 'GUILD'), ('--', ''), ('SN', '1'),
('ID', 'LLM-family F420-associated subfamilies'),
('RQ', '0'),
('EV', 'IPR019910; TIGR03564; sufficient;')]
property_rows_three = [('AC', 'GenProp0092'),
('DE', 'Coenzyme F420 utilization'),
('TP', 'GUILD'), ('--', ''), ('SN', '1'),
('ID',
'LLM-family F420-associated subfamilies'),
('RQ', '0'),
('EV', 'IPR019910; TIGR03565; sufficient;')]
property_one = parse_genome_property(property_rows_one)
property_two = parse_genome_property(property_rows_two)
property_three = parse_genome_property(property_rows_three)
raw_properties = [property_one, property_two, property_three]
cls.tree = GenomePropertiesTree(*raw_properties)
def test_get_step_genome_properties(self):
"""Test that we can get child genome properties directly from a step."""
property_tree = GenomePropertiesTree(*self.properties)
steps = property_tree.root.steps
step_one_genome_properties = steps[0].genome_properties
step_two_genome_properties = steps[1].genome_properties
self.assertEquals(len(step_one_genome_properties), 1)
self.assertEquals(step_one_genome_properties[0].id, 'GenProp0089')
self.assertEquals(len(step_two_genome_properties), 1)
self.assertEquals(step_two_genome_properties[0].id, 'GenProp0092')
def parse_genome_properties_flat_file(genome_property_file):
"""
A parses a genome property flat file.
:param genome_property_file: A genome property file handle object.
:return: A GenomePropertyTree object.
"""
genome_properties = []
current_genome_property_record = []
for line in genome_property_file:
if not line.strip() == '//':
current_genome_property_record.append(create_marker_and_content(line))
else:
collapsed_genome_property_record = unwrap_genome_property_record(current_genome_property_record)
new_genome_property = parse_genome_property(collapsed_genome_property_record)
genome_properties.append(new_genome_property)
current_genome_property_record = []
genome_properties_tree = GenomePropertiesTree(*genome_properties)
return genome_properties_tree
def test_get_consortium_identifiers(self):
"""Test that we can get a set of all consortium identifiers used as evidence by the genome properties tree."""
property_tree = GenomePropertiesTree(*self.properties)
self.assertEqual(property_tree.consortium_identifiers,
{'TIGR03564', 'TIGR03565'})
def test_get_interpro_identifiers(self):
"""Test that we can get a set of all InterPro identifiers used as evidence by the genome properties tree."""
property_tree = GenomePropertiesTree(*self.properties)
self.assertEqual(property_tree.interpro_identifiers,
{'IPR019910', 'IPR019911'})