def __init__(self,
identifier,
title,
url,
description=None,
license_url=None,
data_rights=None,
graph_type=None,
file_handle=None):
if graph_type is None:
self.graph = RDFGraph()
elif graph_type == 'streamed_graph':
self.graph = StreamedGraph(True, file_handle=file_handle)
elif graph_type == 'rdf_graph':
self.graph = RDFGraph()
self.model = Model(self.graph)
self.identifier = ':' + identifier
self.version = None
self.date_issued = None
# The data_accesed value is later used as an object literal of properties such as dct:issued, which needs to conform xsd:dateTime format.
# self.date_accessed = datetime.now().strftime('%Y-%m-%d-%H-%M')
self.date_accessed = datetime.now().strftime('%Y-%m-%dT%H:%M:%S')
self.citation = set()
self.license = license_url
self.model.addType(self.identifier, 'dctypes:Dataset')
self.graph.addTriple(self.identifier, 'dct:title', title, True)
self.graph.addTriple(self.identifier,
'dct:identifier',
identifier,
object_is_literal=True)
self.graph.addTriple(self.identifier, 'foaf:page', url)
# maybe in the future add the logo here:
# schemaorg:logo <http://www.ebi.ac.uk/rdf/sites/ebi.ac.uk.rdf/files/resize/images/rdf/chembl_service_logo-146x48.gif> .
# TODO add the licence info
# FIXME:Temporarily making this in IF statement,
# can revert after all current resources are updated.
if license_url is not None:
self.graph.addTriple(self.identifier, 'dct:license', license_url)
else:
logger.debug('No license provided.')
if data_rights is not None:
self.graph.addTriple(self.identifier,
'dct:rights',
data_rights,
object_is_literal=True)
else:
logger.debug('No rights provided.')
if description is not None:
self.model.addDescription(self.identifier, description)
return
def test_germline_variant_to_disease(self):
self.orphanet.graph = RDFGraph() # Reset graph
self.orphanet.files['disease-gene']['file'] = 'orph-germline.xml'
self.orphanet._process_diseasegene(limit=None)
LOG.debug(
"Reference graph: %s",
self.orphanet.graph.serialize(format="turtle").decode("utf-8")
)
expected_triples = """
MONARCH:ba2ac5d2153c70e2bb98 a OBAN:association ;
RO:0002558 ECO:0000322 ;
OBAN:association_has_object ORPHA:938475 ;
OBAN:association_has_predicate RO:0004013 ;
OBAN:association_has_subject HGNC:30497 .
ENSEMBL:ENSG00000166813 a owl:Class .
HGNC:30497 a owl:Class ;
RO:0004013 ORPHA:938475 ;
oboInOwl:hasExactSynonym "KAS1" ;
owl:equivalentClass ENSEMBL:ENSG00000166813,
ORPHA:268061 .
ORPHA:268061 a owl:Class .
ORPHA:938475 a owl:Class ;
rdfs:label "too much unit testing disorder" .
"""
self.assertTrue(self.test_util.test_graph_equality(
expected_triples, self.orphanet.graph))
return
def setUp(self):
self.test_util = TestUtils()
self.source = GWASCatalog('rdf_graph', True)
self.source.graph = RDFGraph(True)
self.test_data = {
'snp_label': 'rs1329573-?; rs7020413-?; rs3824344-?; rs3758171-?',
'chrom_num': '9;9;9;9',
'chrom_pos': '36998996;37002118;37000690;36997420',
'context':
'intron_variant; intron_variant; intron_variant; intron_variant',
'allele_freq': 'NR',
'trait': 'Intelligence',
'trait_uri': 'http://www.ebi.ac.uk/efo/EFO_0004337',
'pvalue': '0.00000004',
'merged': '0',
'snp_id_current': '',
'mapped_gene': 'PAX5; PAX5; PAX5; PAX5',
'snp_gene_nums': '',
'upstream_gene_num': '107986179',
'downstream_gene_num': '107986180',
'init_sample_desc':
'656 European ancestry individuals from ADHD families',
'replicated_sample_desc': 'NA',
'platform': 'Illumina [795637]',
'pubmed': '22449649'
}
def test_parse(self):
self.source.graph = RDFGraph(True) # Reset graph
self.assertTrue(len(list(self.source.graph)) == 0)
self.source.parse()
triples = """
UNII:46U771ERWK RO:0002606 SNOMED:386761002 ;
rdfs:subClassOf CHEBI:23367 .
SNOMED:386761002 rdfs:label "Local anesthesia" ;
rdfs:subClassOf MONDO:0000001 .
UNII:46U771ERWK biolink:category biolink:ChemicalSubstance .
SNOMED:386761002 biolink:category biolink:Disease .
"""
# dbg
logger.debug(
"Reference graph: %s", self.source.graph.serialize(format="turtle")
.decode("utf-8")
)
self.assertTrue(self.test_util.test_graph_equality(
triples, self.source.graph))
def test_gene_to_disease(self):
self.orphanet.graph = RDFGraph() # Reset graph
self.orphanet.files['disease-gene']['file'] = 'orph-no-variant.xml'
self.orphanet._process_diseasegene(limit=None)
LOG.debug(
"Reference graph: %s",
self.orphanet.graph.serialize(format="turtle") .decode("utf-8")
)
expected_triples = """
MONARCH:bdbeb077e365ddedda20 a OBAN:association ;
RO:0002558 ECO:0000322 ;
OBAN:association_has_object ORPHA:938475 ;
OBAN:association_has_predicate RO:0004015 ;
OBAN:association_has_subject ORPHA:268061 .
ORPHA:268061 RO:0004015 ORPHA:938475 ;
oboInOwl:hasExactSynonym "KAS1" .
ORPHA:938475 a owl:Class ;
rdfs:label "too much unit testing disorder" .
"""
self.assertTrue(self.test_util.test_graph_equality(
expected_triples, self.orphanet.graph))
return
def test_patient_phenotype_model(self):
"""
functional test for _parse_patient_phenotypes()
"""
udp = UDP('rdf_graph', True)
udp.graph = RDFGraph(True)
# test that graph is empty
self.assertTrue(len(list(udp.graph)) == 0)
mock_lines = [
'patient_1\tHP:000001\tyes',
'patient_1\tHP:000002\tno'
]
mock_data = MagicMock()
mock_data.__iter__.return_value = iter(mock_lines)
mock_file = mock_open(mock=mock_data)
udp._parse_patient_phenotypes(mock_file)
triples = """
:patient_1 a foaf:Person ;
rdfs:label "patient_1" ;
RO:0002200 DOID:4,
HP:000001 .
"""
self.assertTrue(self.test_util.test_graph_equality(
triples, udp.graph))
def testSGDParser(self):
sgd = SGD('rdf_graph', True)
sgd.graph = RDFGraph(True)
record = self.test_set_1
sgd.make_association(record)
description = sgd._make_description(record)
triples = """
:MONARCH_ba748c98c0f167739128 a OBAN:association ;
OBO:RO_0002558 OBO:APO_0000020 ;
dcterms:description "{0}";
dcterms:source PMID:21715656 ;
OBAN:association_has_object MONARCH:APO_0000309APO_0000245 ;
OBAN:association_has_predicate OBO:RO_0002200 ;
OBAN:association_has_subject SGD:S000007268 .
SGD:S000007268 rdfs:label "ATP6" ;
RO:0002200 MONARCH:APO_0000309APO_0000245 .
APO:0000020 rdfs:label "classical genetics" .
PMID:21715656 a OBO:IAO_0000311 ;
owl:sameAs SGD_REF:S000145858 .
MONARCH:APO_0000309APO_0000245 rdfs:label "respiratory growth:decreased rate" ;
rdfs:subClassOf UPHENO:0001001 .
""".format(description)
# test exact contents of graph
self.assertTrue(self.test_util.test_graph_equality(triples, sgd.graph))
def test_evidence_model(self):
"""
Functional test for _add_evidence()
"""
impc = IMPC('rdf_graph', True)
impc.graph = RDFGraph(True) # Reset graph
# Test graph is empty
self.assertTrue(len(list(impc.graph)) == 0)
(p_value, percentage_change, effect_size) = self.test_set_1[23:26]
impc._add_evidence(
self.assoc_curie, self.eco_id, p_value, percentage_change, effect_size,
self.study_curie)
triples = """
:MONARCH_test_association SEPIO:0000007 <https://monarchinitiative.org/.well-known/genid/b97a98087df7a99d8a38> .
<https://monarchinitiative.org/.well-known/genid/b97a98087df7a99d8a38> a ECO:0000015 ;
SEPIO:0000084 <https://monarchinitiative.org/.well-known/genid/b41ad2bfd375c9de8888>,
<https://monarchinitiative.org/.well-known/genid/b216606de82749b03956> ;
SEPIO:0000085 <https://monarchinitiative.org/.well-known/genid/study> .
<https://monarchinitiative.org/.well-known/genid/b216606de82749b03956> a OBI:0000175 ;
RO:0002353 <https://monarchinitiative.org/.well-known/genid/study> ;
STATO:0000129 1.637023e-10 .
<https://monarchinitiative.org/.well-known/genid/b41ad2bfd375c9de8888> a STATO:0000085 ;
RO:0002353 <https://monarchinitiative.org/.well-known/genid/study> ;
STATO:0000129 "8.885439E-007" .
"""
self.assertTrue(self.test_util.test_graph_equality(
triples, impc.graph))
def test_parse(self):
for rcv in RCVS:
output_nt = rcv + '.nt'
input_xml = rcv + '.xml.gz'
reference_ttl = TTL_PATH + rcv + '.ttl'
with self.subTest(rcv=rcv):
mock_args = [
"test_clinvar.py", "--inputdir", XML_PATH, "--filename",
input_xml, "--mapfile", MAP_FILE, "--destination", NT_PATH,
"--output", output_nt
]
patch('sys.argv', mock_args).start()
clinvar_parse()
query_graph = RDFGraph()
query_graph.bind_all_namespaces()
query_graph.parse(NT_PATH + output_nt, format='nt')
with open(reference_ttl, 'r') as ref_fh:
ref_graph = "\n".join(ref_fh.readlines())
# debug
LOG.debug(
"Reference graph: %s",
query_graph.serialize(format="turtle").decode("utf-8"))
# Convert output from ClinVar parse to dot then png
dot_file_path = DOT_PATH + rcv + ".dot"
with open(dot_file_path, 'w') as dot_file:
rdf2dot(query_graph, dot_file)
self.assertTrue(
TestUtils.test_graph_equality(ref_graph, query_graph))
def test_graph_equality(turtlish, graph):
"""
:param turtlish: String of triples in turtle
format without prefix header
:param graph: Graph object to test against
:return: Boolean, True if graphs contain same
set of triples
"""
turtle_graph = RDFGraph()
turtle_graph.bind_all_namespaces()
prefixes = "\n".join([
"@prefix {}: <{}> .".format(n[0], n[1])
for n in turtle_graph.namespace_manager.namespaces()
])
turtle_string = prefixes + turtlish
mock_file = io.StringIO(turtle_string)
turtle_graph.parse(mock_file, format="turtle")
turtle_triples = set(list(turtle_graph))
ref_triples = set(list(graph))
equality = turtle_triples == ref_triples
if not equality:
LOG.warning(
"Triples do not match\n"
"\tLeft hand difference: %s\n"
"\tRight hand difference: %s",
sorted(turtle_triples - ref_triples),
sorted(ref_triples - turtle_triples))
return equality
def test_assertion_model(self):
"""
Functional test for _add_study_provenance()
"""
impc = IMPC('rdf_graph', True)
impc.graph = RDFGraph(True)
self.assertTrue(len(list(impc.graph)) == 0)
impc._add_assertion_provenance(self.assoc_curie, self.evidence_curie)
triples = """
MONARCH:test_association SEPIO:0000015 <https://monarchinitiative.org/.well-known/genid/bf92df374a884963e805> .
<https://monarchinitiative.org/.well-known/genid/bf92df374a884963e805> a SEPIO:0000001 ;
SEPIO:0000018 <https://www.mousephenotype.org/> ;
SEPIO:0000111 <https://monarchinitiative.org/.well-known/genid/evidence> .
<https://www.mousephenotype.org/> a foaf:organization ;
rdfs:label "International Mouse Phenotyping Consortium" .
"""
# dbg
LOG.info(
"Assertion graph:\n %s\n", impc.graph.serialize(
format="turtle").decode("utf-8")
)
self.assertTrue(self.test_util.test_graph_equality(triples, impc.graph))
def testEnsemblReactomeParser(self):
reactome = Reactome('rdf_graph', True)
reactome.graph = RDFGraph(True)
self.assertTrue(len(list(reactome.graph)) == 0)
eco_map = Reactome.get_eco_map(Reactome.map_files['eco_map'])
(gene, pathway_id, pathway_iri, pathway_label, go_ecode,
species_name) = self.test_set_1
reactome._add_component_pathway_association(eco_map, gene, 'ENSEMBL',
pathway_id, 'REACT',
pathway_label, go_ecode)
triples = """
ENSEMBL:ENSBTAP00000013354 RO:0002331 REACT:R-BTA-3000480 .
:MONARCH_b582c188b7ec20016206 a OBAN:association ;
OBO:RO_0002558 ECO:0000501 ;
OBAN:association_has_object REACT:R-BTA-3000480 ;
OBAN:association_has_predicate RO:0002331 ;
OBAN:association_has_subject ENSEMBL:ENSBTAP00000013354 .
REACT:R-BTA-3000480 a owl:Class ;
rdfs:label "Scavenging by Class A Receptors" ;
rdfs:subClassOf GO:0009987,
PW:0000001 .
"""
self.assertTrue(
self.test_util.test_graph_equality(triples, reactome.graph))
def test_germline_lof_variant_to_disease(self):
self.orphanet.graph = RDFGraph() # Reset graph
self.orphanet.files['disease-gene']['file'] = 'orph-germline-lof.xml'
self.orphanet._process_diseasegene(limit=None)
logger.debug(
"Reference graph: %s",
self.orphanet.graph.serialize(format="turtle").decode("utf-8"))
expected_triples = """
MONARCH:b53dada0eb229a75e705 OBAN:association ;
RO:0002558 ECO:0000322 ;
OBAN:association_has_object Orphanet:938475 ;
OBAN:association_has_predicate RO:0003303 ;
OBAN:association_has_subject <https://monarchinitiative.org/.well-known/genid/ba0884fb61004110> .
Orphanet:268061 a owl:Class ;
rdfs:label "KS1" ;
dc:description "kinesin family member 7" ;
oboInOwl:hasExactSynonym "KAS1" ;
rdfs:subClassOf SO:0001217 .
<https://monarchinitiative.org/.well-known/genid/ba0884fb61004110> a GENO:0000002 ;
rdfs:label "germline loss of function variant of KS1" ;
GENO:0000418 Orphanet:268061 ;
RO:0003303 Orphanet:938475 ;
:MONARCH_anonymous true ;
:has_cell_origin GENO:0000900 ;
:has_functional_consequence SO:0002054 .
Orphanet:938475 a owl:Class ;
rdfs:label "too much unit testing disorder" .
"""
self.assertTrue(
self.test_util.test_graph_equality(expected_triples,
self.orphanet.graph))
def test_germline_lof_variant_to_disease(self):
self.orphanet.graph = RDFGraph() # Reset graph
self.orphanet.files['disease-gene']['file'] = 'orph-germline-lof.xml'
self.orphanet._process_diseasegene(limit=None)
LOG.debug(
"Reference graph: %s",
self.orphanet.graph.serialize(format="turtle").decode("utf-8"))
expected_triples = """
MONARCH:b9ad1b0c562ad4db3f1e a OBAN:association ;
RO:0002558 ECO:0000322 ;
OBAN:association_has_object ORPHA:938475 ;
OBAN:association_has_predicate RO:0004012 ;
OBAN:association_has_subject ORPHA:268061 .
ORPHA:268061 RO:0004012 ORPHA:938475 ;
oboInOwl:hasExactSynonym "KAS1" .
ORPHA:938475 a owl:Class ;
rdfs:label "too much unit testing disorder" .
ECO:0000322 biolink:category biolink:EvidenceType .
ORPHA:268061 biolink:category biolink:Gene .
ORPHA:268061 biolink:category biolink:Genotype .
ORPHA:938475 biolink:category biolink:Disease .
MONARCH:b9ad1b0c562ad4db3f1e biolink:category biolink:Association .
"""
self.assertTrue(
self.test_util.test_graph_equality(expected_triples,
self.orphanet.graph))
return
def test_gene_to_disease(self):
self.orphanet.graph = RDFGraph() # Reset graph
self.orphanet.files['disease-gene']['file'] = 'orph-no-variant.xml'
self.orphanet._process_diseasegene(limit=None)
logger.debug(
"Reference graph: %s",
self.orphanet.graph.serialize(format="turtle").decode("utf-8"))
expected_triples = """
MONARCH:b64684a0ea6ae59fdb09 a OBAN:association ;
RO:0002558 ECO:0000322 ;
OBAN:association_has_object Orphanet:938475 ;
OBAN:association_has_predicate RO:0003304 ;
OBAN:association_has_subject Orphanet:268061 .
Orphanet:268061 a owl:Class ;
rdfs:label "KS1" ;
RO:0003304 Orphanet:938475 ;
dc:description "kinesin family member 7" ;
oboInOwl:hasExactSynonym "KAS1" ;
rdfs:subClassOf SO:0001217 .
Orphanet:938475 a owl:Class ;
rdfs:label "too much unit testing disorder" .
"""
self.assertTrue(
self.test_util.test_graph_equality(expected_triples,
self.orphanet.graph))
def setUp(self):
self.test_util = TestUtils()
self.source = GWASCatalog('rdf_graph', True)
self.source.graph = RDFGraph(True) # Reset graph
self.source.graph.bind_all_namespaces()
self.test_data = {
'snp_label': 'rs1491921-C',
'chrom_num': '5',
'chrom_pos': '21259029',
'context': 'intergenic_variant',
'allele_freq': '0.013',
'trait': 'Diisocyanate-induced asthma',
'trait_uri': 'http://www.ebi.ac.uk/efo/EFO_0006995, http://www.ebi.ac.uk/efo/EFO_0003949',
'pvalue': '0.0000007',
'merged': '0',
'snp_id_current': '1491921',
'mapped_gene': 'LOC102723561 - GUSBP1',
'snp_gene_nums': '',
'upstream_gene_num': '107986179',
'downstream_gene_num': '107986180',
'init_sample_desc': '74 European ancestry cases, 824 European ancestry controls',
'replicated_sample_desc': 'NA',
'platform': 'Illumina [1556551]',
'pubmed': '25918132'
}
def testFakeDataSet1(self):
string_db = StringDB('rdf_graph', True)
string_db.graph = RDFGraph(True)
self.assertEqual(len(string_db.graph), 0)
ensembl = Ensembl('rdf_graph', True)
prot_map = ensembl.fetch_protein_gene_map('9606')
[prot_map.update({k: ['ENSEMBL:' + prot_map[k]]}) for k in prot_map.keys()]
print("Finished fetching ENSP IDs, fetched {} proteins".format(len(prot_map)))
# just looking
# for key in prot_map:
# if string_db.graph.curie_regexp.match(prot_map[key]) is None:
# print("INVALID curie for %s from %s", prot_map[key], key)
dataframe = pd.DataFrame(data=self.test_set_1, columns=self.columns)
string_db._process_protein_links(dataframe, prot_map, '9606')
# g1 <interacts with> g2
triples = """
ENSEMBL:ENSG00000001626 RO:0002434 ENSEMBL:ENSG00000004059 .
"""
self.assertTrue(self.test_util.test_graph_equality(triples, string_db.graph))
def test_snp_trait_association(self):
"""
test the _add_variant_trait_association
:return:
"""
efo_ontology = RDFGraph()
logger.info("Loading EFO ontology in separate rdf graph")
efo_ontology.parse(self.source.files['efo']['url'], format='xml')
efo_ontology.bind_all_namespaces()
logger.info("Finished loading EFO ontology")
variant_curie, variant_type = \
self.source._get_curie_and_type_from_id(self.test_data['snp_label'])
description = self.source._make_description(
self.test_data['trait'], self.test_data['init_sample_desc'],
self.test_data['replicated_sample_desc'],
self.test_data['platform'], self.test_data['pvalue'])
self.source._add_variant_trait_association(variant_curie,
self.test_data['trait_uri'],
efo_ontology,
self.test_data['pubmed'],
description)
sparql_query = """
SELECT ?snp
WHERE {{
<https://monarchinitiative.org/MONARCH_b46cdf48950cb00d> a OBAN:association ;
dc:description "{}" ;
OBO:RO_0002558 OBO:ECO_0000213 ;
dc:source PMID:25918132 ;
OBAN:association_has_object EFO:0003949 ;
OBAN:association_has_predicate OBO:RO_0002326 ;
OBAN:association_has_subject ?snp .
<https://monarchinitiative.org/MONARCH_70a05d8eb1c3d4b0> a OBAN:association ;
OBO:RO_0002558 OBO:ECO_0000213 ;
dc:source PMID:25918132 ;
OBAN:association_has_object EFO:0006995 ;
OBAN:association_has_predicate OBO:RO_0002326 ;
OBAN:association_has_subject ?snp .
EFO:0003949 a owl:Class ;
rdfs:label "eye color"^^xsd:string ;
rdfs:subClassOf UPHENO:0001001 .
?snp OBO:RO_0002326 EFO:0003949,
EFO:0006995 .
PMID:25918132 a OBO:IAO_0000013 .
}}
""".format(description)
sparql_output = self.source.graph.query(sparql_query)
# Test that query passes and returns one row
results = list(sparql_output)
expected = [(URIRef(self.source.graph._getNode("dbSNP:rs1491921")), )]
self.assertEqual(results, expected)
def test_snp_trait_association(self):
"""
test the _add_variant_trait_association
:return:
"""
self.assertTrue(len(list(self.source.graph)) == 0)
efo_ontology = RDFGraph()
LOG.info("Loading EFO ontology in separate rdf graph")
efo_ontology.parse(self.source.files['efo']['url'], format='xml')
efo_ontology.bind_all_namespaces()
LOG.info("Finished loading EFO ontology")
variant_curie, variant_type = self.source._get_curie_and_type_from_id(
self.test_data['snp_label'])
description = self.source._make_description(
self.test_data['trait'], self.test_data['init_sample_desc'],
self.test_data['replicated_sample_desc'],
self.test_data['platform'], self.test_data['pvalue'])
self.source._add_variant_trait_association(
variant_curie, self.test_data['trait_uri'], efo_ontology,
self.test_data['pubmed'], description)
triples = """
MONARCH:bffc7a930c08cc8fe931 a OBAN:association ;
dc:description "{0}" ;
OBO:RO_0002558 OBO:ECO_0000213 ;
dc:source PMID:25918132 ;
OBAN:association_has_object EFO:0003949 ;
OBAN:association_has_predicate RO:0003304 ;
OBAN:association_has_subject dbSNP:rs1491921 .
MONARCH:bff9b97458d67ed7f517 a OBAN:association ;
dc:description "{0}" ;
OBO:RO_0002558 OBO:ECO_0000213 ;
dc:source PMID:25918132 ;
OBAN:association_has_object EFO:0006995 ;
OBAN:association_has_predicate RO:0003304 ;
OBAN:association_has_subject dbSNP:rs1491921 .
EFO:0003949 a owl:Class ;
rdfs:label "eye color"^^xsd:string ;
rdfs:subClassOf UPHENO:0001001 .
dbSNP:rs1491921 RO:0003304 EFO:0003949,
EFO:0006995 .
PMID:25918132 a OBO:IAO_0000013 .
""".format(description)
# dbg
# LOG.debug(
# "Reference graph: %s",
# self.source.graph.serialize(format="turtle").decode("utf-8"))
self.assertTrue(self.test_util.test_graph_equality(triples, self.source.graph))
def test_provenance_model(self):
"""
Functional test for _add_study_provenance()
"""
impc = IMPC('rdf_graph', True)
impc.graph = RDFGraph(True)
self.assertTrue(len(list(impc.graph)) == 0)
(phenotyping_center, colony) = self.test_set_1[2:4]
(project_fullname, pipeline_name, pipeline_stable_id,
procedure_stable_id, procedure_name, parameter_stable_id,
parameter_name) = self.test_set_1[12:19]
(statistical_method, resource_name) = self.test_set_1[26:28]
impc._add_study_provenance(phenotyping_center, colony,
project_fullname, pipeline_name,
pipeline_stable_id, procedure_stable_id,
procedure_name, parameter_stable_id,
parameter_name, statistical_method,
resource_name)
# dbg
LOG.info("Provenance graph as turtle:\n%s\n",
impc.graph.serialize(format="turtle").decode("utf-8"))
triples = """
<https://monarchinitiative.org/.well-known/genid/bdd05a8ca155ddaf415e> a OBI:0000471 ;
BFO:0000051 OBO:STATO_0000076,
<https://www.mousephenotype.org/impress/protocol/175/15> ;
BFO:0000050 IMPRESS-procedure:15 ,
<http://www.sanger.ac.uk/science/data/mouse-genomes-project> ;
SEPIO:0000114 <https://www.mousephenotype.org/impress/parameterontologies/1867/91> ;
SEPIO:0000017 <http://www.sanger.ac.uk/> .
<https://monarchinitiative.org/.well-known/genid/b0b26361b8687b5ad9ef> a owl:NamedIndividual ;
rdfs:label "MEFW" .
<http://www.sanger.ac.uk/> a foaf:organization ;
rdfs:label "WTSI" .
<http://www.sanger.ac.uk/science/data/mouse-genomes-project> a VIVO:Project ;
rdfs:label "Wellcome Trust Sanger Institute Mouse Genetics Project" .
<https://www.mousephenotype.org/impress/parameterontologies/1867/91> a owl:NamedIndividual ;
rdfs:label "Number of ribs right (X-ray)" .
IMPRESS-procedure:15 a owl:NamedIndividual ;
rdfs:label "MGP Select Pipeline" .
<https://www.mousephenotype.org/impress/protocol/175/15> a owl:NamedIndividual ;
rdfs:label "X-ray" .
"""
# dbg
LOG.debug("Reference graph: %s",
impc.graph.serialize(format="turtle").decode("utf-8"))
self.assertTrue(self.test_util.test_graph_equality(
triples, impc.graph))
def setUp(self):
self.test_util = TestUtils()
self.source = CTD('rdf_graph', True)
self.source.graph = RDFGraph(True)
self.test_row = [
'Nicotine', 'D009538', '', 'TOBACCO ADDICTION, SUSCEPTIBILITY TO',
'OMIM:188890', 'therapeutic', '', '', '', '12345|56789'
]
return
def test_variant_model(self):
"""
functional test for _parse_patient_variants()
"""
udp = UDP('rdf_graph', True)
udp.graph = RDFGraph(True)
# test that graph is empty
self.assertTrue(len(list(udp.graph)) == 0)
data = ['patient_1',
'family_1',
'1',
'HG19',
'155230432',
'G',
'A',
'Maternal',
'Biallelic',
'Non-synonymous;DOWNSTREAM',
'CLK2',
'',
'',
'',
'',
'',
'',
'',
'Compound heterozygous',
'Heterozygous',
'',
'0.002747253',
'']
test_data = "\t".join(data)
mock_lines = [test_data]
mock_data = MagicMock()
mock_data.__iter__.return_value = iter(mock_lines)
mock_file = mock_open(mock=mock_data)
udp._parse_patient_variants(mock_file)
triples = """
:patient_1 GENO:0000222 <https://monarchinitiative.org/.well-known/genid/ba5f377fc8c95d4a6d7a> .
<https://monarchinitiative.org/.well-known/genid/b41e8da0787b45e24c4f> a SO:0001059 ;
rdfs:label "hg19chr1(CLK2):g.155230432G>A" ;
GENO:0000418 HGNC:2069 ;
RO:0002162 NCBITaxon:9606 ;
owl:sameAs dbSNP:rs11557757 .
<https://monarchinitiative.org/.well-known/genid/ba5f377fc8c95d4a6d7a> a GENO:0000000 ;
rdfs:label "patient_1 genotype" ;
GENO:0000382 <https://monarchinitiative.org/.well-known/genid/b41e8da0787b45e24c4f> .
"""
self.assertTrue(self.test_util.test_graph_equality(triples, udp.graph))
def test_unmapped_disease_assoc_type(self):
"""
Test that a gene disease type that we have
not mapped in translationtable/orphanet.yaml
raises a ValueError
"""
self.orphanet.graph = RDFGraph() # Reset graph
self.orphanet.files['disease-gene']['file'] = 'orph-no-mapping.xml'
self.assertRaises(
ValueError, lambda: self.orphanet._process_diseasegene(limit=None))
def setUp(self):
self.graph = RDFGraph()
self.curie_map = curie_map.get()
self.genotype = Genotype(self.graph)
self.cutil = CurieUtil(self.curie_map)
self.test_cat_pred = self.cutil.get_uri(blv.terms['category'])
self.test_cat_genotype_category = self.cutil.get_uri(
blv.terms['Genotype'])
self.test_cat_background_category = self.cutil.get_uri(
blv.terms['PopulationOfIndividualOrganisms'])
def testFakeDataSet2(self):
"""
Dataset contains a deprecated protein ID
that we expect if filtered out by ensembl biomart
We test that this returns an empty graph
:return:
"""
string_db = StringDB('rdf_graph', True)
string_db.graph = RDFGraph()
self.assertEqual(len(string_db.graph), 0)
dataframe = pd.DataFrame(data=self.test_set_2, columns=self.columns)
string_db._process_protein_links(dataframe, self.protein_list, 9606)
self.assertEqual(len(string_db.graph), 0)
def test_sex_specificity_model(self):
self.mgi.graph = RDFGraph(True) # Reset graph
self.mgi._process_evidence_view(limit=None)
logger.debug("Reference graph: %s",
self.mgi.graph.serialize(format="turtle").decode("utf-8"))
expected_triples = """
:association RO:0002558 ECO:0000006 ;
dc:source J:74619 ;
:has_sex_specificity PATO:0000384 .
J:74619 a IAO:0000310 .
"""
self.assertTrue(
self.test_util.test_graph_equality(expected_triples,
self.mgi.graph))
def read_graph_from_turtle_file(self, f):
"""
This will read the specified file into a graph. A simple parsing test.
:param f:
:return:
"""
vg = RDFGraph()
p = os.path.abspath(f)
logger.info("Testing reading turtle file from %s", p)
vg.parse(f, format="turtle")
logger.info('Found %s graph nodes in %s', len(vg), p)
self.assertTrue(len(vg) > 0, "No nodes found in " + p)
return
def setUp(self):
self.graph = RDFGraph()
this_curie_map = curie_map.get()
self.cutil = CurieUtil(this_curie_map)
# stuff to make test triples
self.test_cat_subj = "http://www.google.com"
self.test_cat_default_pred = self.cutil.get_uri("biolink:category")
self.test_cat_nondefault_pred = self.cutil.get_uri("rdf:type")
self.test_cat_default_category = self.cutil.get_uri(
"biolink:NamedThing")
self.test_cat_nondefault_category = self.cutil.get_uri("biolink:Gene")
self.test_cat_type = self.cutil.get_uri("rdf:type")
self.test_cat_class = self.cutil.get_uri("rdf:class")
def setUp(self):
g = RDFGraph()
self.model = Model(g)
this_curie_map = curie_map.get()
self.cutil = CurieUtil(this_curie_map)
# stuff to make test triples
self.test_cat_subj_curie = "MGI:1234"
self.test_cat_subj = self.cutil.get_uri("MGI:1234")
self.test_cat_default_pred = self.cutil.get_uri("biolink:category")
self.test_named_indiv = self.cutil.get_uri("owl:NamedIndividual")
self.test_label_pred = self.cutil.get_uri("rdfs:label")
self.test_label = "some label"
self.test_comment_IRI = self.cutil.get_uri("rdfs:comment")
self.test_comment = 'bonus eruptus'
def test_graph_equality(turtlish, graph):
"""
:param turtlish: file path or string of triples in turtle
format without prefix header
:param graph: Graph object to test against
:return: Boolean, True if graphs contain same
set of triples
"""
turtle_graph = RDFGraph()
turtle_graph.bind_all_namespaces()
prefixes = "\n".join([
"@prefix {}: <{}> .".format(n[0], n[1])
for n in turtle_graph.namespace_manager.namespaces()
])
headless_ttl = ''
try:
if Path(turtlish).exists():
headless_ttl = Path(turtlish).read_text()
else:
raise OSError
except OSError:
if isinstance(turtlish, str):
headless_ttl = turtlish
else:
raise ValueError("turtlish must be filepath or string")
turtle_string = prefixes + headless_ttl
mock_file = io.StringIO(turtle_string)
turtle_graph.parse(mock_file, format="turtle")
TestUtils.remove_ontology_axioms(graph)
turtle_triples = set(list(turtle_graph))
ref_triples = set(list(graph))
equality = turtle_triples == ref_triples
if not equality:
LOG.warning(
"Triples do not match\n"
"\tLeft hand difference: %s\n"
"\tRight hand difference: %s",
sorted(turtle_triples - ref_triples),
sorted(ref_triples - turtle_triples))
return equality