def get_properties_from_input(file, input_format):
input_graph = ConjunctiveGraph()
input_graph.parse(file, format=input_format)
# collapse to single list
property_set = list()
for row in input_graph.predicates():
property_set.append(row)
return set(property_set)
def get_properties_from_input(file, input_format):
input_graph = ConjunctiveGraph()
input_graph.parse(file, format=input_format)
# collapse to single list
property_set = set()
for row in input_graph.predicates():
property_set.add(row)
return property_set
# or: primer.add((myNS['pat'], myNS['age'], long(24))) # Now, with just that, lets see how the system # recorded *way* too many details about what # you just asserted as fact. # from pprint import pprint pprint(list(primer)) # just think .whatever((s, p, o)) # here we report on what we know pprint(list(primer.subjects())) pprint(list(primer.predicates())) pprint(list(primer.objects())) # and other things that make sense # what do we know about pat? pprint(list(primer.predicate_objects(myNS.pat))) # who is what age? pprint(list(primer.subject_objects(myNS.age))) # Okay, so lets now work with a bigger # dataset from the example, and start # with a fresh new graph. primer = ConjunctiveGraph()
def make_property_graph(properties, args):
graph = ConjunctiveGraph()
output_graph = ConjunctiveGraph()
GH = 'https://raw.githubusercontent.com'
OBO = 'https://purl.obolibrary.org/obo'
ontologies = [
OBO + '/sepio.owl',
OBO + '/geno.owl',
OBO + '/iao.owl',
OBO + '/ero.owl',
OBO + '/pco.owl',
OBO + '/xco.owl',
OBO + '/ro.owl',
GH + '/jamesmalone/OBAN/master/ontology/oban_core.ttl',
]
for ontology in ontologies:
print("parsing: " + ontology)
try:
graph.parse(ontology, format=rdflib_util.guess_format(ontology))
except SAXParseException as e:
logger.error(e)
logger.error('Retrying: ' + ontology)
graph.parse(ontology, format="turtle")
except OSError as e: # URLError:
# simple retry
logger.error(e)
logger.error('Retrying: ' + ontology)
graph.parse(ontology, format=rdflib_util.guess_format(ontology))
# Get object properties
output_graph = add_property_to_graph(
graph.subjects(RDF['type'], OWL['ObjectProperty']), output_graph,
OWL['ObjectProperty'], properties)
# Get annotation properties
output_graph = add_property_to_graph(
graph.subjects(RDF['type'], OWL['AnnotationProperty']), output_graph,
OWL['AnnotationProperty'], properties)
# Get data properties
output_graph = add_property_to_graph(
graph.subjects(RDF['type'], OWL['DatatypeProperty']), output_graph,
OWL['DatatypeProperty'], properties)
# Hardcoded properties
output_graph.add(
(URIRef('https://monarchinitiative.org/MONARCH_cliqueLeader'),
RDF['type'], OWL['AnnotationProperty']))
output_graph.add(
(URIRef('https://monarchinitiative.org/MONARCH_anonymous'),
RDF['type'], OWL['AnnotationProperty']))
# Check monarch data triple
data_url = "https://data.monarchinitiative.org/ttl/{0}".format(
re.sub(r".*/", "", args.input))
new_url = "https://data.monarchinitiative.org/ttl/{0}".format(
re.sub(r".*/", "", args.output))
if (URIRef(data_url), RDF.type, OWL['Ontology']) in output_graph:
output_graph.remove(URIRef(data_url), RDF.type, OWL['Ontology'])
output_graph.add((URIRef(new_url), RDF.type, OWL['Ontology']))
for row in output_graph.predicates(DC['source'],
OWL['AnnotationProperty']):
if row == RDF['type']:
output_graph.remove(
(DC['source'], RDF['type'], OWL['AnnotationProperty']))
output_graph.add((DC['source'], RDF['type'], OWL['ObjectProperty']))
return output_graph
def make_property_graph(properties, args):
graph = ConjunctiveGraph()
output_graph = ConjunctiveGraph()
ontologies = [
'https://raw.githubusercontent.com/monarch-initiative/SEPIO-ontology/master/src/ontology/sepio.owl',
'https://raw.githubusercontent.com/monarch-initiative/GENO-ontology/develop/src/ontology/geno.owl',
'http://purl.obolibrary.org/obo/ro.owl',
'http://purl.obolibrary.org/obo/iao.owl',
'http://purl.obolibrary.org/obo/ero.owl',
'https://raw.githubusercontent.com/jamesmalone/OBAN/master/ontology/oban_core.ttl',
'http://purl.obolibrary.org/obo/pco.owl',
'http://purl.obolibrary.org/obo/xco.owl'
]
for ontology in ontologies:
print("parsing: " + ontology)
try:
graph.parse(ontology, format=rdflib_util.guess_format(ontology))
except SAXParseException as e:
logger.error(e)
logger.error('Retrying: ' + ontology)
graph.parse(ontology, format="turtle")
except OSError as e: # URLError:
# simple retry
logger.error(e)
logger.error('Retrying: ' + ontology)
graph.parse(ontology, format=rdflib_util.guess_format(ontology))
# Get object properties
output_graph = add_property_to_graph(
graph.subjects(RDF['type'], OWL['ObjectProperty']),
output_graph, OWL['ObjectProperty'], properties)
# Get annotation properties
output_graph = add_property_to_graph(
graph.subjects(RDF['type'], OWL['AnnotationProperty']),
output_graph, OWL['AnnotationProperty'], properties)
# Get data properties
output_graph = add_property_to_graph(
graph.subjects(RDF['type'], OWL['DatatypeProperty']),
output_graph, OWL['DatatypeProperty'], properties)
# Hardcoded properties
output_graph.add(
(URIRef('https://monarchinitiative.org/MONARCH_cliqueLeader'),
RDF['type'], OWL['AnnotationProperty']))
output_graph.add(
(URIRef('https://monarchinitiative.org/MONARCH_anonymous'),
RDF['type'], OWL['AnnotationProperty']))
# Check monarch data triple
data_url = "https://data.monarchinitiative.org/ttl/{0}".format(
re.sub(r".*/", "", args.input))
new_url = "https://data.monarchinitiative.org/ttl/{0}".format(
re.sub(r".*/", "", args.output))
if (URIRef(data_url), RDF.type, OWL['Ontology']) in output_graph:
output_graph.remove(URIRef(data_url), RDF.type, OWL['Ontology'])
output_graph.add((URIRef(new_url), RDF.type, OWL['Ontology']))
for row in output_graph.predicates(
DC['source'], OWL['AnnotationProperty']):
if row == RDF['type']:
output_graph.remove(
(DC['source'], RDF['type'], OWL['AnnotationProperty']))
output_graph.add((DC['source'], RDF['type'], OWL['ObjectProperty']))
return output_graph
# Now, with just that, lets see how the system # recorded *way* too many details about what # you just asserted as fact. # from pprint import pprint pprint(list(primer)) # just think .whatever((s, p, o)) # here we report on what we know pprint(list(primer.subjects())) pprint(list(primer.predicates())) pprint(list(primer.objects())) # and other things that make sense # what do we know about pat? pprint(list(primer.predicate_objects(myNS.pat))) # who is what age? pprint(list(primer.subject_objects(myNS.age))) # Okay, so lets now work with a bigger # dataset from the example, and start # with a fresh new graph.
