def parse(self, limit=None):
zfin_parser = ZFIN(self.graph_type, self.are_bnodes_skized)
model = Model(self.graph)
zp_file = '/'.join((self.rawdir, self.files['zpmap']['file']))
g2p_file = '/'.join((self.rawdir, self.files['g2p_clean']['file']))
zfin_parser.zp_map = zfin_parser._load_zp_mappings(zp_file)
with open(g2p_file, 'r', encoding="utf8") as csvfile:
filereader = csv.reader(csvfile, delimiter='\t', quotechar='\"')
for row in filereader:
(internal_id, symbol, gene_id, subterm1_id, subterm1_label,
pc_rel_id, pc_rel_label, superterm1_id, superterm1_label,
quality_id, quality_name, modifier, subterm2_id,
subterm2_label, pc_rel2_id, pc_rel2_id, superterm2_id,
superterm2_label, fish_id, fish_label, start_stage, end_stage,
environment, pub_id, figure_id, unknown_field) = row
zp_id = zfin_parser._map_sextuple_to_phenotype(
superterm1_id, subterm1_id, quality_id, superterm2_id,
subterm2_id, modifier)
gene_curie = "ZFIN:{0}".format(gene_id)
model.makeLeader(gene_curie)
pub_curie = "ZFIN:{0}".format(pub_id)
if zp_id:
assoc = G2PAssoc(self.graph, self.name, gene_curie, zp_id)
if pub_id:
reference = Reference(self.graph, pub_curie,
Reference.ref_types['document'])
reference.addRefToGraph()
assoc.add_source(pub_curie)
assoc.add_evidence('ECO:0000059')
assoc.add_association_to_graph()
def parse(self, limit=None):
zfin_parser = ZFIN(self.graph_type, self.are_bnodes_skized)
model = Model(self.graph)
zp_file = '/'.join((self.rawdir, self.files['zpmap']['file']))
g2p_file = '/'.join((self.rawdir, self.files['g2p_clean']['file']))
zfin_parser.zp_map = zfin_parser._load_zp_mappings(zp_file)
with open(g2p_file, 'r', encoding="utf8") as csvfile:
filereader = csv.reader(csvfile, delimiter='\t', quotechar='\"')
for row in filereader:
(internal_id, symbol, gene_id, subterm1_id, subterm1_label,
pc_rel_id, pc_rel_label, superterm1_id, superterm1_label,
quality_id, quality_name, modifier, subterm2_id,
subterm2_label, pc_rel2_id, pc_rel2_id, superterm2_id,
superterm2_label, fish_id, fish_label, start_stage, end_stage,
environment, pub_id, figure_id, unknown_field) = row
zp_id = zfin_parser._map_sextuple_to_phenotype(
superterm1_id, subterm1_id, quality_id, superterm2_id,
subterm2_id, modifier)
gene_curie = "ZFIN:{0}".format(gene_id)
model.makeLeader(gene_curie)
pub_curie = "ZFIN:{0}".format(pub_id)
if zp_id:
assoc = G2PAssoc(self.graph, self.name, gene_curie, zp_id)
if pub_id:
reference = Reference(self.graph, pub_curie,
Reference.ref_types['document'])
reference.addRefToGraph()
assoc.add_source(pub_curie)
assoc.add_evidence('ECO:0000059')
assoc.add_association_to_graph()
class ZFINTestCase(SourceTestCase):
def setUp(self):
self.source = ZFIN('rdf_graph', True)
self.source.settestonly(True)
self._setDirToSource()
def tearDown(self):
self.source = None
def test_mapping_of_phenotypes_to_zp_ids(self):
"""
test that code correctly uses zp_map to map phenotypes to zp ids
:return:
"""
mapping_file = "./tests/resources/zfin/zp-mapping-test-map.txt"
pheno_file = "./tests/resources/zfin/zp-mapping-test-phenotype.txt"
self.source.zp_map = self.source._load_zp_mappings(mapping_file)
pheno_dat = open(pheno_file).read().split('\t')
(fish_num, fish_name, start_stage_id, start_stage_name, end_stage_id,
end_stage_name, subterm1_id, subterm1_name, postcomp1_rel_id,
postcomp1_rel_name, superterm1_id, superterm1_name, quality_id,
quality_name, modifier, subterm2_id, subterm2_name, postcomp2_rel_id,
postcomp2_rel_name, superterm2_id, superterm2_name, pub_id, env_id) = \
pheno_dat
self.assertEqual(
self.source._map_octuple_to_phenotype(
subterm1_id, postcomp1_rel_id, superterm1_id, quality_id,
subterm2_id, postcomp2_rel_id, superterm2_id, "abnormal"),
'ZP:0022140')
def test_load_zp_mappings(self):
"""
test correct loading of zp mappings file and construction of zp_map
"""
if self.source is not None:
try:
zp_map = self.source._load_zp_mappings(
"./tests/resources/zfin/zp-mapping-test.txt")
self.assertIsInstance(
zp_map, dict, "_load_zp_mappings() didn't return dict!")
self.assertTrue(
len(zp_map) == 1,
"_load_zp_mappings() didn't return exactly one thing!")
self.assertDictEqual(
zp_map, {
'MONARCH:b308a8f1c67793a56d16': {
'post_composed_relationship_id_1': 'BFO:0000050',
'post_composed_relationship_id_2': 'BFO:0000050',
'quality_id': 'PATO:0001453',
'subterm1_id': 'ZFA:0009114',
'subterm2_id': 'GO:0005927',
'superterm1_id': 'ZFA:0001056',
'superterm2_id': 'ZFA:0001056',
'zp_id': 'ZP:0002959',
'modifier': 'PATO:0000460'
}
},
"_load_zp_mappings() " + "didn't return what I expected!")
except Exception as t_except:
LOGGER.error(t_except)
def test_make_zpkey(self):
"""
test that _make_zpkey returns correct id
"""
if self.source is not None:
try:
dummy_args = list(map(str, list(range(1,
9)))) # 1 - 8 as strings
expected_key = self.source.make_id("_".join(dummy_args))
self.assertEqual(self.source._make_zpkey(*dummy_args),
expected_key)
self.assertEqual(
self.source._make_zpkey(['0'] * 8),
self.source._make_zpkey([''] * 8),
"_make_zpkey() doesn't seem to be replacing empty " +
"strings with zeros before making key," +
"this might cause zp_map lookup issues")
except Exception as t_except:
LOGGER.error(t_except)
def test_genotype_labels(self):
"""
test that genotype label is set correctly after parse()
"""
if self.source is not None:
test_resource_dir = "../../tests/resources/zfin/"
self.source.files['fish_components']['file'] = test_resource_dir + \
"genotype-label-test-fish_components_fish.txt"
self.source.files['backgrounds']['file'] = test_resource_dir + \
"genotype-label-test-genotype_backgrounds.txt"
self.source.files['geno']['file'] = test_resource_dir + \
"genotype-label-test-genotype_features.txt"
self.source.parse()
this_iri = URIRef("http://zfin.org/ZDB-GENO-070228-3")
expect_genotype_label = "shha<sup>tbx392/tbx392</sup> (AB)"
self.assertEqual(str(self.source.testgraph.label(this_iri, None)),
expect_genotype_label)
def parse(self, limit=None):
zfin_parser = ZFIN(self.graph_type, self.are_bnodes_skized)
model = Model(self.graph)
src_key = 'zpmap' # keep same-as zfin.files[key]
zfin_parser.zp_map = zfin_parser._load_zp_mappings(src_key)
src_key = 'g2p_clean'
raw = '/'.join((self.rawdir, self.files[src_key]['file']))
LOG.info("Processing clean Geno to Pheno from file: %s", raw)
col = self.files[src_key]['columns']
collen = len(col)
with open(raw, 'r', encoding="utf8") as csvfile:
reader = csv.reader(csvfile, delimiter='\t', quotechar='\"')
for row in reader:
if len(row) != collen:
LOG.warning('Row: %i has unexpected format',
reader.line_num)
# internal_id = row[col.index('ID')]
# symbol = row[col.index('Gene Symbol')]
gene_id = row[col.index('Gene ID')]
subterm1_id = row[col.index(
'Affected Structure or Process 1 subterm ID')]
# subterm1_label = row[col.index(
# 'Affected Structure or Process 1 subterm Name')]
pc_rel_id = row[col.index(
'Post-composed Relationship ID')].strip()
# pc_rel_label = row[col.index('Post-composed Relationship Name')]
superterm1_id = row[col.index(
'Affected Structure or Process 1 superterm ID')].strip()
# superterm1_label = row[col.index(
# 'Affected Structure or Process 1 superterm Name')]
quality_id = row[col.index('Phenotype Keyword ID')].strip()
# quality_name = row[col.index('Phenotype Keyword Name')]
modifier = row[col.index('Phenotype Tag')].strip()
subterm2_id = row[col.index(
'Affected Structure or Process 2 subterm ID')].strip()
# subterm2_label = row[col.index(
# 'Affected Structure or Process 2 subterm name')]
pc_rel2_id = row[col.index(
'Post-composed Relationship (rel) ID')]
# pc_rel2_label = row[col.index(
# 'Post-composed Relationship (rel) Name')]
superterm2_id = row[col.index(
'Affected Structure or Process 2 superterm ID')].strip()
# superterm2_label = row[col.index(
# 'Affected Structure or Process 2 superterm name')]
# fish_id = row[col.index('Fish ID')]
# fish_label = row[col.index('Fish Display Name')]
start_stage = row[col.index('Start Stage ID')]
# end_stage = row[col.index('End Stage ID')]
# environment = row[col.index('Fish Environment ID')]
pub_id = row[col.index('Publication ID')].strip()
# figure_id = row[col.index('Figure ID')]
if modifier != 'abnormal':
LOG.warning(
"skipping phenotype with modifier %s != abnormal ",
modifier)
continue
zp_id = zfin_parser._map_octuple_to_phenotype(
subterm1_id, pc_rel_id, superterm1_id, quality_id,
subterm2_id, pc_rel2_id, superterm2_id, modifier)
gene_curie = "ZFIN:{0}".format(gene_id)
model.makeLeader(gene_curie)
pub_curie = "ZFIN:{0}".format(pub_id)
if zp_id:
assoc = G2PAssoc(self.graph, self.name, gene_curie, zp_id)
if pub_id:
reference = Reference(self.graph, pub_curie,
self.globaltt['document'])
reference.addRefToGraph()
assoc.add_source(pub_curie)
assoc.add_evidence(
self.globaltt['experimental phenotypic evidence'])
assoc.add_association_to_graph()
