def test_save1():
"""
save tests
"""
src_path = "tests/resources/monarch/biogrid_test.ttl"
t = ObanRdfTransformer()
t.parse(src_path, input_format="turtle")
save(t.graph, 'biogrid')
def test_validator_rdf():
"""
use test files
"""
cwd = os.path.abspath(os.path.dirname(__file__))
resdir = os.path.join(cwd, 'resources')
src_path = os.path.join(resdir, 'monarch', 'biogrid_test.ttl')
t = ObanRdfTransformer()
t.parse(src_path, input_format="turtle")
validator = Validator()
validator.validate(t.graph)
write_errors(validator)
def test_mapping():
"""
create a random graph and save it in different formats
"""
G = nx.MultiDiGraph()
N = 100
E = N * 3
mapping = {}
for i in range(0,N+1):
nid = curie(i)
mapping[nid] = mapped_curie(i)
G.add_node(nid, label="node {}".format(i))
for i in range(1,E):
s = random_curie(N)
o = random_curie(N)
G.add_edge(o,s,predicate='related_to')
print('Nodes={}'.format(len(G.nodes())))
mapper.map_graph(G, mapping)
print("Mapped..")
count = 0
for nid in G.nodes():
src = G.node[nid]['source_curie']
assert nid.startswith("Y:")
assert src.startswith("X:")
count += 1
if count > 5:
break
print("Saving tsv")
w = PandasTransformer(G)
w.save("target/maptest.tar")
w = ObanRdfTransformer(G)
w.save("target/maptest.ttl")
def test_save_ttl():
t = ObanRdfTransformer()
t.parse(os.path.join(resource_dir, 'hpoa_test.ttl'))
assert t.graph.number_of_nodes() == 10
assert t.graph.number_of_edges() == 5
t.save(os.path.join(target_dir, 'hpoa_test_export.ttl'))
assert os.path.exists(os.path.join(target_dir, 'hpoa_test_export.ttl'))
def test_load_ttl():
t = ObanRdfTransformer()
t.parse(os.path.join(resource_dir, 'hpoa_test.ttl'))
assert t.graph.number_of_nodes() == 10
assert t.graph.number_of_edges() == 5
import pprint
pprint.pprint([x for x in t.graph.nodes(data=True)])
pprint.pprint([x for x in t.graph.edges(data=True)])
n1 = t.graph.nodes(data=True)['HP:0000007']
assert n1['id'] == 'HP:0000007'
assert n1['provided_by'] == ['hpoa_test.ttl']
assert n1['category'] == ['biolink:NamedThing']
e1 = t.graph.get_edge_data('Orphanet:93262', 'HP:0000505')
data = e1.popitem()
assert data[1]['subject'] == 'Orphanet:93262'
assert data[1]['edge_label'] == 'biolink:has_phenotype'
assert data[1]['object'] == 'HP:0000505'
assert data[1]['relation'] == 'RO:0002200'
assert data[1]['provided_by'] == ['hpoa_test.ttl']
assert data[1]['has_evidence'] == 'ECO:0000304'
def test_load():
"""
load and save tests
"""
cwd = os.path.abspath(os.path.dirname(__file__))
src_path = os.path.join(cwd, 'resources', 'monarch', 'biogrid_test.ttl')
tpath = os.path.join(cwd, 'target')
os.makedirs(tpath, exist_ok=True)
tg_path = os.path.join(tpath, "test_output.ttl")
# execute ObanRdfTransformer's parse and save function
t = ObanRdfTransformer()
t.parse(src_path, input_format="turtle")
t.save(tg_path, output_format="turtle")
t.report()
w1 = PandasTransformer(t.graph)
w1.save(os.path.join(tpath, 'biogrid-e.csv'), type='e')
w1.save(os.path.join(tpath, 'biogrid-n.csv'), type='n')
# read again the source, test graph
src_graph = rdflib.Graph()
src_graph.parse(src_path, format="turtle")
# read again the dumped target graph
tg_graph = rdflib.Graph()
tg_graph.parse(tg_path, format="turtle")
# compare subgraphs from the source and the target graph.
OBAN = Namespace('http://purl.org/oban/')
for a in src_graph.subjects(RDF.type, OBAN.association):
oban_src_graph = rdflib.Graph()
oban_src_graph += src_graph.triples((a, None, None))
oban_tg_graph = rdflib.Graph()
oban_tg_graph += tg_graph.triples((a, None, None))
# see they are indeed identical (isomorphic)
if not oban_src_graph.isomorphic(oban_tg_graph):
raise RuntimeError('The subgraphs whose subject is ' + str(a) +
' are not isomorphic ones.')
w2 = GraphMLTransformer(t.graph)
w2.save(os.path.join(tpath, "x1n.graphml"))
w3 = JsonTransformer(t.graph)
w3.save(os.path.join(tpath, "x1n.json"))
def test_load():
"""
load TTL and save as CSV
"""
input_file = os.path.join(resource_dir, 'monarch/biogrid_test.ttl')
output_file = os.path.join(target_dir, 'test_output.ttl')
t = ObanRdfTransformer()
t.parse(input_file, input_format="turtle")
t.report()
t.save(output_file, output_format="turtle")
output_archive_file = os.path.join(target_dir, 'biogrid_test')
pt = PandasTransformer(t.graph)
pt.save(output_archive_file)
# read again the source, test graph
src_graph = rdflib.Graph()
src_graph.parse(input_file, format="turtle")
# read again the dumped target graph
target_graph = rdflib.Graph()
target_graph.parse(output_file, format="turtle")
# compare subgraphs from the source and the target graph.
OBAN = Namespace('http://purl.org/oban/')
for a in src_graph.subjects(RDF.type, OBAN.association):
oban_src_graph = rdflib.Graph()
oban_src_graph += src_graph.triples((a, None, None))
oban_tg_graph = rdflib.Graph()
oban_tg_graph += target_graph.triples((a, None, None))
# see they are indeed identical (isomorphic)
if not oban_src_graph.isomorphic(oban_tg_graph):
print(
'The subgraphs whose subject is {} are not isomorphic'.format(
a))
# w2 = GraphMLTransformer(t.graph)
# w2.save(os.path.join(tpath, "x1n.graphml"))
w3 = JsonTransformer(t.graph)
w3.save(os.path.join(target_dir, "biogrid_test.json"))
def test_load():
"""
load tests
"""
t = ObanRdfTransformer()
t.parse("tests/resources/monarch/biogrid_test.ttl")
t.report()
w1 = PandasTransformer(t)
w1.save('target/biogrid-e.csv', type='e')
w1.save('target/biogrid-n.csv', type='n')
w2 = GraphMLTransformer(t)
w2.save("target/x1n.graphml")
w3 = JsonTransformer(t)
w3.save("target/x1n.json")
def test_load():
"""
create a random graph and save it in different formats
"""
G = nx.MultiDiGraph()
# Adjust this to test for scalability
N = 1000
E = N * 3
for i in range(1, N):
G.add_node(curie(i), label="node {}".format(i))
for i in range(1, E):
s = random_curie(N)
o = random_curie(N)
G.add_edge(o, s)
print('Nodes={}'.format(len(G.nodes())))
rename_all(G)
print("Saving tsv")
w = PandasTransformer(source=G)
w.save("target/random.tar")
print("Saving ttl")
w = ObanRdfTransformer(source=G)
w.save("target/random.ttl")
#o = ObanRdfTransformer()
#o.add_ontology('data/mondo.owl')
#o.add_ontology('data/hp.owl')
#o.add_ontology('data/go.owl')
#o.add_ontology('data/so.owl')
#o.add_ontology('data/ordo.owl')
from rdflib import URIRef
t = HgncRdfTransformer()
t.parse('data/hgnc.ttl')
t = JsonTransformer(t)
t.save('hgnc.json')
quit()
t = ObanRdfTransformer()
t.ontologies = o.ontologies
t.parse('data/orphanet.ttl')
t = JsonTransformer(t)
t.save('orphanet.json')
t = ObanRdfTransformer()
t.ontologies = o.ontologies
t.parse('data/hpoa.ttl')
t = JsonTransformer(t)
t.save('hpoa.json')
t = ObanRdfTransformer()
t.ontologies = o.ontologies
t.parse('data/omim.ttl')
t = JsonTransformer(t)