class TestGOModule(unittest.TestCase): def setUp(self): logging.basicConfig(filename=None, level="ERROR", format='%(asctime)s - %(name)s - %(levelname)s: %(message)s') logger.info("Starting DataManager tests") self.this_dir = os.path.split(__file__)[0] self.conf_parser = GenedescConfigParser(os.path.join(self.this_dir, os.path.pardir, "tests", "config_test.yml")) self.df = DataManager(do_relations=None, go_relations=["subClassOf", "BFO:0000050"]) logger.info("Loading go ontology from file") self.df.load_ontology_from_file(ontology_type=DataType.GO, ontology_url="file://" + os.path.join( self.this_dir, "data", "go_gd_test.obo"), ontology_cache_path=os.path.join(self.this_dir, "cache", "go_gd_test.obo"), config=self.conf_parser) logger.info("Loading go associations from file") self.df.load_associations_from_file(associations_type=DataType.GO, associations_url="file://" + os.path.join( self.this_dir, "data", "gene_association_1.7.wb.partial"), associations_cache_path=os.path.join(self.this_dir, "cache", "gene_association_1.7.wb.partial"), config=self.conf_parser) def test_ontology_exists(self): self.assertTrue(self.df.go_ontology is not None) self.assertTrue(any(parent == "GO:0009987" for parent in self.df.go_ontology.parents("GO:0000075"))) def test_annotations_exist(self): self.assertTrue(self.df.go_associations is not None) self.assertTrue(len(self.df.get_annotations_for_gene( gene_id="WB:WBGene00000001", annot_type=DataType.GO, include_obsolete=False, include_negative_results=False, priority_list=self.conf_parser.get_annotations_priority(module=Module.GO))) > 0) def test_rename_terms(self): self.assertTrue(all(len(self.df.go_ontology.search(term)) == 0 for term in list( self.conf_parser.get_module_property(module=Module.GO, prop=ConfigModuleProperty.RENAME_TERMS).keys()))) def test_exclude_terms(self): test_annot = self.df.get_annotations_for_gene("WB:WBGene00000001", annot_type=DataType.GO) self.assertTrue(all([annot["object"]["id"] != "GO:0008286" for annot in test_annot])) def test_download_gz_file(self): test_file = self.df._get_cached_file(cache_path=os.path.join(self.this_dir, "cache", "c_elegans.PRJNA13758.WS273.geneIDs.txt.gz"), file_source_url="file://" + os.path.join( self.this_dir, "data", "c_elegans.PRJNA13758.WS273.geneIDs.txt.gz")) self.assertTrue(test_file == os.path.join(self.this_dir, "cache", "c_elegans.PRJNA13758.WS273.geneIDs.txt")) def test_gene_data_functions(self): self.df.set_gene_data(gene_data=[Gene("1", "gene1", True, False), Gene("2", "gene2", False, True), Gene("3", "gene3", False, False), Gene("4", "gene4", True, True)]) self.assertTrue(len([g for g in self.df.get_gene_data(include_dead_genes=False, include_pseudo_genes=False)]) == 1) self.assertTrue(len([g for g in self.df.get_gene_data(include_dead_genes=True, include_pseudo_genes=False)]) == 2) self.assertTrue(len([g for g in self.df.get_gene_data(include_dead_genes=False, include_pseudo_genes=True)]) == 2) self.assertTrue(len([g for g in self.df.get_gene_data(include_dead_genes=True, include_pseudo_genes=True)]) == 4) def test_get_human_gene_props(self): human_gene_props = self.df.get_human_gene_props() self.assertTrue(len(human_gene_props) > 0) def test_get_ensembl_hgnc_ids_map(self): ensembl_hgnc_ids_map = self.df.get_ensembl_hgnc_ids_map() self.assertTrue(len(ensembl_hgnc_ids_map) > 0) def test_set_ontology(self): ontology = OntologyFactory().create() for i in range(4): ontology.add_node(i, 'node' + str(i)) ontology.add_parent(1, 0) ontology.add_parent(2, 0) ontology.add_parent(3, 0) self.df.set_ontology(ontology_type=DataType.GO, ontology=ontology, config=self.conf_parser) self.assertTrue(list(self.df.go_ontology.nodes()) == list(ontology.nodes())) 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_and_process_data(self): # create gene descriptions data manager and load common data context_info = ContextInfo() data_manager = DataFileManager(context_info.config_file_location) #go_onto_config = data_manager.get_config('GO') go_annot_config = data_manager.get_config('GAF') #do_onto_config = data_manager.get_config('DOID') go_annot_sub_dict = {sub.get_data_provider(): sub for sub in go_annot_config.get_sub_type_objects()} this_dir = os.path.split(__file__)[0] gd_config = GenedescConfigParser(os.path.join(this_dir, os.pardir, os.pardir, "gene_descriptions.yml")) gd_data_manager = DataManager(do_relations=None, go_relations=["subClassOf", "BFO:0000050"]) gd_data_manager.set_ontology(ontology_type=DataType.GO, ontology=self.get_ontology(data_type=DataType.GO), config=gd_config) gd_data_manager.set_ontology(ontology_type=DataType.DO, ontology=self.get_ontology(data_type=DataType.DO), config=gd_config) # generate descriptions for each MOD for prvdr in [sub_type.get_data_provider().upper() \ for sub_type in self.data_type_config.get_sub_type_objects()]: gd_config_mod_specific = copy.deepcopy(gd_config) if prvdr == "WB": gd_config_mod_specific.config["expression_sentences_options"][ "remove_children_if_parent_is_present"] = True self.logger.info("Generating gene descriptions for %s", prvdr) data_provider = prvdr if prvdr != "HUMAN" else "RGD" json_desc_writer = DescriptionsWriter() go_annot_path = "file://" + os.path.join(os.getcwd(), "tmp", go_annot_sub_dict[prvdr].file_to_download) gd_data_manager.load_associations_from_file( associations_type=DataType.GO, associations_url=go_annot_path, associations_cache_path=os.path.join(os.getcwd(), "tmp", "gd_cache", "go_annot_" + prvdr + ".gaf"), config=gd_config_mod_specific) gd_data_manager.set_associations(associations_type=DataType.DO, associations=self.get_disease_annotations_from_db( data_provider=data_provider, gd_data_manager=gd_data_manager, logger=self.logger), config=gd_config_mod_specific) if prvdr in EXPRESSION_PRVD_SUBTYPE_MAP: gd_data_manager.set_ontology(ontology_type=DataType.EXPR, ontology=self.get_ontology(data_type=DataType.EXPR, provider=prvdr), config=gd_config_mod_specific) gd_data_manager.set_associations( associations_type=DataType.EXPR, associations=self.get_expression_annotations_from_db(data_provider=data_provider, gd_data_manager=gd_data_manager, logger=self.logger), config=gd_config_mod_specific) commit_size = self.data_type_config.get_neo4j_commit_size() generators = self.get_generators(prvdr, gd_data_manager, gd_config_mod_specific, json_desc_writer) query_template_list = [ [self.gene_descriptions_query_template, commit_size, "genedescriptions_data_" + prvdr + ".csv"] ] query_and_file_list = self.process_query_params(query_template_list) CSVTransactor.save_file_static(generators, query_and_file_list) Neo4jTransactor.execute_query_batch(query_and_file_list) self.save_descriptions_report_files(data_provider=prvdr, json_desc_writer=json_desc_writer, context_info=context_info, gd_data_manager=gd_data_manager)