def test_get_common_ancestors(self):
self.load_go_ontology()
generator = OntologySentenceGenerator(gene_id="WB:WBGene00000912", module=Module.GO,
data_manager=self.df, config=self.conf_parser)
node_ids = generator.terms_groups[('P', '')]["EXPERIMENTAL"]
common_ancestors = get_all_common_ancestors(node_ids, generator.ontology)
self.assertTrue(len(common_ancestors) > 0, "Common ancestors not found")
associations = [association for subj_associations in self.df.go_associations.associations_by_subj.values() for
association in subj_associations]
associations.append(DataManager.create_annotation_record(source_line="", gene_id="WB:WBGene00003931",
gene_symbol="", gene_type="gene", taxon_id="",
object_id="GO:0043055", qualifiers="", aspect="P",
ecode="EXP", references="", prvdr="WB", date=""))
associations.append(DataManager.create_annotation_record(source_line="", gene_id="WB:WBGene00003931",
gene_symbol="", gene_type="gene", taxon_id="",
object_id="GO:0061065", qualifiers="", aspect="P",
ecode="EXP", references="", prvdr="WB", date=""))
associations.append(DataManager.create_annotation_record(source_line="", gene_id="WB:WBGene00003931",
gene_symbol="", gene_type="gene", taxon_id="",
object_id="GO:0043054", qualifiers="", aspect="P",
ecode="EXP", references="", prvdr="WB", date=""))
associations.append(DataManager.create_annotation_record(source_line="", gene_id="WB:WBGene00003931",
gene_symbol="", gene_type="gene", taxon_id="",
object_id="GO:0043053", qualifiers="", aspect="P",
ecode="EXP", references="", prvdr="WB", date=""))
self.df.go_associations = AssociationSetFactory().create_from_assocs(assocs=associations,
ontology=self.df.go_ontology)
self.conf_parser.config["go_sentences_options"]["exclude_terms"].append("GO:0040024")
generator = OntologySentenceGenerator(gene_id="WB:WBGene00003931", module=Module.GO,
data_manager=self.df, config=self.conf_parser)
node_ids = generator.terms_groups[('P', '')]["EXPERIMENTAL"]
common_ancestors = get_all_common_ancestors(node_ids, generator.ontology)
self.assertTrue("GO:0040024" not in common_ancestors, "Common ancestors contain blacklisted term")
def test_set_associations(self):
associations = []
associations.append(DataManager.create_annotation_record("", "1", "a", "protein_coding", "001", "GO:0019901",
"", "F", "EXP", None, "WB", ""))
associations.append(DataManager.create_annotation_record("", "2", "b", "protein_coding", "001", "GO:0005515",
"", "F", "EXP", None, "WB", ""))
assocs = AssociationSetFactory().create_from_assocs(assocs=associations, ontology=self.df.go_ontology)
self.df.set_associations(associations_type=DataType.GO, associations=assocs, config=self.conf_parser)
self.assertTrue(self.df.go_associations)
def test_remap_associations(self):
associations = []
associations.append(DataManager.create_annotation_record("", "1", "a", "protein_coding", "001", "GO:0018996",
"", "F", "EXP", None, "WB", ""))
assocs = AssociationSetFactory().create_from_assocs(assocs=associations, ontology=self.df.go_ontology)
self.df.set_associations(associations_type=DataType.GO, associations=assocs, config=self.conf_parser)
self.assertEqual(self.df.go_associations.associations_by_subj["1"][0]["object"]["id"], "GO:0042303")
def _load_expression_cluster_file(
self,
file_cache_path,
file_url,
load_into_data,
add_to_expression_ontology_annotations: bool = False):
expr_clust_file = self._get_cached_file(cache_path=file_cache_path,
file_source_url=file_url)
header = True
associations = []
terms_ids_map = {}
if add_to_expression_ontology_annotations:
associations = [
association for subj_associations in
self.expression_associations.associations_by_subj.values()
for association in subj_associations
]
terms_replacement_regex = self.config.get_module_property(
module=Module.EXPRESSION, prop=ConfigModuleProperty.RENAME_TERMS)
for line in open(expr_clust_file):
if not header:
linearr = line.strip().split("\t")
load_into_data[linearr[0]] = linearr[1:]
load_into_data[
linearr[0]][2] = WBDataManager.get_replaced_terms_arr(
load_into_data[linearr[0]][2].split(","),
terms_replacement_regex)
if load_into_data[linearr[0]] and load_into_data[
linearr[0]][3]:
load_into_data[linearr[0]][3] = [
word.replace(" study", "").replace(" analysis", "")
for word in load_into_data[linearr[0]][3].split(",")
]
if add_to_expression_ontology_annotations:
for term in load_into_data[linearr[0]][2]:
if term not in terms_ids_map:
term_ids = self.expression_ontology.resolve_names(
[term])
if term_ids:
terms_ids_map[term] = term_ids[0]
else:
terms_ids_map[term] = None
if term in terms_ids_map and terms_ids_map[term]:
associations.append(
DataManager.create_annotation_record(
line, "WB:" + linearr[0], "", "gene", "",
terms_ids_map[term], ["Enriched"], "A",
"IDA", "", "", ""))
else:
header = False
if add_to_expression_ontology_annotations:
self.set_associations(
DataType.EXPR,
associations=AssociationSetFactory().create_from_assocs(
assocs=associations, ontology=self.expression_ontology),
config=self.config)
def get_associations(gene_id, term_ids, qualifiers, aspect, ecode):
return [DataManager.create_annotation_record(source_line="", gene_id=gene_id, gene_symbol="", gene_type="gene",
taxon_id="", object_id=term_id, qualifiers=qualifiers,
aspect=aspect, ecode=ecode, references="", prvdr="WB",
date="") for term_id in term_ids]
def load_associations_from_file(
self,
associations_type: DataType,
associations_url: str,
associations_cache_path: str,
config: GenedescConfigParser,
association_additional_url: str = None,
association_additional_cache_path: str = None) -> None:
logger.info("Loading associations from file")
if associations_type == DataType.GO:
super().load_associations_from_file(
associations_type=associations_type,
associations_url=associations_url,
associations_cache_path=associations_cache_path,
config=config)
elif associations_type == DataType.EXPR:
associations = []
file_path = self._get_cached_file(
cache_path=associations_cache_path,
file_source_url=associations_url)
for line in open(file_path):
if not line.strip().startswith("!"):
linearr = line.strip().split("\t")
if self.expression_ontology.node(linearr[4]):
gene_id = linearr[0] + ":" + linearr[1]
qualifiers = linearr[3].split("|")
if len(
qualifiers
) == 0 or "Partial" in qualifiers or "Certain" in qualifiers:
qualifiers = ["Verified"]
associations.append(
DataManager.create_annotation_record(
line, gene_id, linearr[2], linearr[11],
linearr[12], linearr[4], qualifiers,
linearr[8], linearr[6], linearr[5].split("|"),
linearr[14], linearr[13]))
self.expression_associations = AssociationSetFactory(
).create_from_assocs(assocs=associations,
ontology=self.expression_ontology)
self.expression_associations = self.remove_blacklisted_annotations(
association_set=self.expression_associations,
ontology=self.expression_ontology,
terms_blacklist=config.get_module_property(
module=Module.EXPRESSION,
prop=ConfigModuleProperty.EXCLUDE_TERMS))
elif associations_type == DataType.DO:
self.do_associations = AssociationSetFactory().create_from_assocs(
assocs=GafParser().parse(file=self._get_cached_file(
cache_path=associations_cache_path,
file_source_url=associations_url),
skipheader=True),
ontology=self.do_ontology)
if association_additional_cache_path and association_additional_url:
associations = []
for subj_associations in self.do_associations.associations_by_subj.values(
):
for association in subj_associations:
if association["evidence"]["type"] == "IEA":
associations.append(association)
file_path = self._get_cached_file(
cache_path=association_additional_cache_path,
file_source_url=association_additional_url)
header = True
for line in open(file_path):
if not line.strip().startswith("!"):
if not header:
linearr = line.strip().split("\t")
if self.do_ontology.node(
linearr[10]) and linearr[16] != "IEA":
gene_ids = [linearr[2]]
if linearr[1] == "allele":
gene_ids = linearr[4].split(",")
for gene_id in gene_ids:
associations.append(
DataManager.create_annotation_record(
line, gene_id, linearr[3],
linearr[1], linearr[0],
linearr[10], linearr[9].split("|"),
"D", linearr[16],
linearr[18].split("|"),
linearr[20], linearr[19]))
else:
header = False
self.do_associations = AssociationSetFactory(
).create_from_assocs(assocs=associations,
ontology=self.do_ontology)
self.do_associations = self.remove_blacklisted_annotations(
association_set=self.do_associations,
ontology=self.do_ontology,
terms_blacklist=config.get_module_property(
module=Module.DO_EXPERIMENTAL,
prop=ConfigModuleProperty.EXCLUDE_TERMS))
def test_set_covering_with_ontology(self):
self.load_do_ontology()
self.conf_parser.config["do_via_orth_sentences_options"]["trimming_algorithm"] = "ic"
self.conf_parser.config["do_via_orth_sentences_options"]["max_num_terms"] = 5
associations = [DataManager.create_annotation_record(source_line="", gene_id="MGI:88452",
gene_symbol="", gene_type="gene", taxon_id="",
object_id="DOID:0080028", qualifiers="", aspect="D",
ecode="ISS", references="", prvdr="WB", date=""),
DataManager.create_annotation_record(source_line="", gene_id="MGI:88452",
gene_symbol="", gene_type="gene", taxon_id="",
object_id="DOID:0080056", qualifiers="", aspect="D",
ecode="ISS", references="", prvdr="WB", date=""),
DataManager.create_annotation_record(source_line="", gene_id="MGI:88452",
gene_symbol="", gene_type="gene", taxon_id="",
object_id="DOID:14789", qualifiers="", aspect="D",
ecode="ISS", references="", prvdr="WB", date=""),
DataManager.create_annotation_record(source_line="", gene_id="MGI:88452",
gene_symbol="", gene_type="gene", taxon_id="",
object_id="DOID:0080026", qualifiers="", aspect="D",
ecode="ISS", references="", prvdr="WB", date=""),
DataManager.create_annotation_record(source_line="", gene_id="MGI:88452",
gene_symbol="", gene_type="gene", taxon_id="",
object_id="DOID:14415", qualifiers="", aspect="D",
ecode="ISS", references="", prvdr="WB", date=""),
DataManager.create_annotation_record(source_line="", gene_id="MGI:88452",
gene_symbol="", gene_type="gene", taxon_id="",
object_id="DOID:0080045", qualifiers="", aspect="D",
ecode="ISS", references="", prvdr="WB", date=""),
DataManager.create_annotation_record(source_line="", gene_id="MGI:88452",
gene_symbol="", gene_type="gene", taxon_id="",
object_id="DOID:3371", qualifiers="", aspect="D",
ecode="ISS", references="", prvdr="WB", date=""),
DataManager.create_annotation_record(source_line="", gene_id="MGI:88452",
gene_symbol="", gene_type="gene", taxon_id="",
object_id="DOID:8886", qualifiers="", aspect="D",
ecode="ISS", references="", prvdr="WB", date=""),
DataManager.create_annotation_record(source_line="", gene_id="MGI:88452",
gene_symbol="", gene_type="gene", taxon_id="",
object_id="DOID:674", qualifiers="", aspect="D",
ecode="ISS", references="", prvdr="WB", date=""),
DataManager.create_annotation_record(source_line="", gene_id="MGI:88452",
gene_symbol="", gene_type="gene", taxon_id="",
object_id="DOID:5614", qualifiers="", aspect="D",
ecode="ISS", references="", prvdr="WB", date=""),
DataManager.create_annotation_record(source_line="", gene_id="MGI:88452",
gene_symbol="", gene_type="gene", taxon_id="",
object_id="DOID:11830", qualifiers="", aspect="D",
ecode="ISS", references="", prvdr="WB", date=""),
DataManager.create_annotation_record(source_line="", gene_id="MGI:88452",
gene_symbol="", gene_type="gene", taxon_id="",
object_id="DOID:8398", qualifiers="", aspect="D",
ecode="ISS", references="", prvdr="WB", date=""),
DataManager.create_annotation_record(source_line="", gene_id="MGI:88452",
gene_symbol="", gene_type="gene", taxon_id="",
object_id="DOID:2256", qualifiers="", aspect="D",
ecode="ISS", references="", prvdr="WB", date=""),
DataManager.create_annotation_record(source_line="", gene_id="MGI:88452",
gene_symbol="", gene_type="gene", taxon_id="",
object_id="DOID:5327", qualifiers="", aspect="D",
ecode="ISS", references="", prvdr="WB", date=""),
DataManager.create_annotation_record(source_line="", gene_id="MGI:88452",
gene_symbol="", gene_type="gene", taxon_id="",
object_id="DOID:1123", qualifiers="", aspect="D",
ecode="ISS", references="", prvdr="WB", date="")]
self.df.do_associations = AssociationSetFactory().create_from_assocs(assocs=associations,
ontology=self.df.do_ontology)
generator = OntologySentenceGenerator(gene_id="MGI:88452", module=Module.DO_ORTHOLOGY,
data_manager=self.df, config=self.conf_parser)
sentences = generator.get_module_sentences(
config=self.conf_parser, aspect='D', qualifier='', merge_groups_with_same_prefix=True,
keep_only_best_group=True, high_priority_term_ids=["DOID:0080028", "DOID:0080056", "DOID:14789",
"DOID:0080026", "DOID:14415", "DOID:0080045"])
print(sentences.get_description())