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_compose_sentence(self):
this_dir = os.path.split(__file__)[0]
conf_parser = GenedescConfigParser(
os.path.join(this_dir, os.path.pardir, "tests", "config_test.yml"))
sentence = compose_sentence(
prefix="Is expressed in",
additional_prefix="several processes, including",
term_names=["cell", "tail", "head", "male"],
postfix="based on experimental observation",
config=conf_parser,
ancestors_with_multiple_children={"head"},
rename_cell=True,
put_anatomy_male_at_end=True)
self.assertTrue("cell" not in sentence)
self.assertTrue("and in male" in sentence)
sentence = compose_sentence(
prefix="Is expressed in",
additional_prefix="several processes, including",
term_names=["cell"],
postfix="based on experimental observation",
config=conf_parser,
rename_cell=True)
self.assertTrue(
sentence ==
"Is expressed widely based on experimental observation")
def setUp(self):
logging.basicConfig(
filename=None,
level="INFO",
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"))
def load_do_ontology(self):
logger.info("Starting Ontology Tools 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)
logger.info("Loading do ontology from file")
logging.basicConfig(filename=None, level="ERROR", format='%(asctime)s - %(name)s - %(levelname)s: %(message)s')
self.df.load_ontology_from_file(ontology_type=DataType.DO, ontology_url="file://" + os.path.join(
self.this_dir, "data", "doid.obo"),
ontology_cache_path=os.path.join(self.this_dir, "cache", "doid.obo"),
config=self.conf_parser)
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, "config_test_wb.yml"))
self.df = WBDataManager(do_relations=None,
go_relations=["subClassOf", "BFO:0000050"],
config=self.conf_parser,
species="c_elegans")
def main():
parser = argparse.ArgumentParser(
description="Generate gene descriptions for wormbase")
parser.add_argument("-c",
"--config-file",
metavar="config_file",
dest="config_file",
type=str,
default="config.yml",
help="configuration file. Default ./config.yaml")
parser.add_argument(
"-C",
"--use-cache",
dest="use_cache",
action="store_true",
default=False,
help=
"Use cached source files from cache_location specified in config file. Download them from "
"raw_file_source (configured in config file) if not yet cached")
parser.add_argument(
"-l",
"--log-file",
metavar="log_file",
dest="log_file",
type=str,
default=None,
help="path to the log file to generate. Default ./genedescriptions.log"
)
parser.add_argument(
"-L",
"--log-level",
dest="log_level",
choices=['DEBUG', 'INFO', 'WARNING', 'ERROR', 'CRITICAL'],
help="set the logging level")
parser.add_argument("-t",
"--textpressoapi-token",
metavar="textpresso_token",
dest="textpresso_token",
type=str,
help="Texpresso api token")
parser.add_argument("-o",
"--output-formats",
metavar="output_formats",
dest="output_formats",
type=str,
nargs="+",
default=["ace", "txt", "json", "tsv"],
help="file formats to generate. Accepted values "
"are: ace, txt, json, tsv")
args = parser.parse_args()
conf_parser = GenedescConfigParser(args.config_file)
logging.basicConfig(filename=args.log_file,
level=args.log_level,
format='%(asctime)s - %(name)s - %(levelname)s:'
'%(message)s',
force=True)
logger = logging.getLogger("WB Gene Description Pipeline")
organisms_list = conf_parser.get_wb_organisms_to_process()
human_genes_props = DataManager.get_human_gene_props()
api_manager = APIManager(textpresso_api_token=args.textpresso_token)
for organism in organisms_list:
logger.info("Processing organism " + organism)
species = conf_parser.get_wb_organisms_info()
dm, sister_df, df_agr = load_data(organism=organism,
conf_parser=conf_parser)
desc_writer = DescriptionsWriter()
desc_writer.overall_properties.species = organism
desc_writer.overall_properties.release_version = conf_parser.get_wb_release(
)[0:-1] + str(int(conf_parser.get_wb_release()[-1]) + 1)
desc_writer.overall_properties.date = datetime.date.today().strftime(
"%B %d, %Y")
for gene in dm.get_gene_data():
logger.debug("Generating description for gene " + gene.name)
gene_desc = GeneDescription(gene_id=gene.id,
config=conf_parser,
gene_name=gene.name,
add_gene_name=False)
selected_orthologs, orth_sent = get_best_orthologs_and_sentence(
dm=dm,
orth_fullnames=dm.orth_fullnames,
human_genes_props=human_genes_props,
gene_desc=gene_desc,
api_manager=api_manager,
config=conf_parser)
set_gene_ontology_module(dm=dm,
conf_parser=conf_parser,
gene_desc=gene_desc,
gene=gene)
set_tissue_expression_sentence(dm=dm,
gene=gene,
conf_parser=conf_parser,
gene_desc=gene_desc)
if not gene_desc.description:
set_expression_cluster_sentence(dm=dm,
conf_parser=conf_parser,
gene_desc=gene_desc,
gene=gene,
api_manager=api_manager)
set_disease_module(df=dm,
conf_parser=conf_parser,
gene=gene,
gene_desc=gene_desc)
if not gene_desc.go_description:
set_information_poor_sentence(
orth_fullnames=dm.orth_fullnames,
selected_orthologs=selected_orthologs,
conf_parser=conf_parser,
human_df_agr=df_agr,
gene_desc=gene_desc,
dm=dm,
gene=gene)
gene_desc.set_or_extend_module_description_and_final_stats(
module=Module.ORTHOLOGY, description=orth_sent)
if "main_sister_species" in species[organism] and species[organism]["main_sister_species"] and \
dm.get_best_orthologs_for_gene(gene.id, orth_species_full_name=[dm.sister_sp_fullname],
sister_species_data_fetcher=sister_df,
ecode_priority_list=["EXP", "IDA", "IPI", "IMP", "IGI", "IEP", "HTP",
"HDA", "HMP", "HGI", "HEP"])[0]:
set_sister_species_sentence(
dm=dm,
sister_sp_fullname=dm.sister_sp_fullname,
sister_df=sister_df,
species=species,
organism=organism,
gene_desc=gene_desc,
conf_parser=conf_parser,
gene=gene)
desc_writer.add_gene_desc(gene_desc)
logger.info("All genes processed for " + organism)
date_prefix = datetime.date.today().strftime("%Y%m%d")
if "json" in args.output_formats:
logger.info("Writing descriptions to json")
desc_writer.write_json(os.path.join(
conf_parser.get_out_dir(),
date_prefix + "_" + organism + ".json"),
include_single_gene_stats=True,
data_manager=dm)
if "txt" in args.output_formats:
logger.info("Writing descriptions to txt")
desc_writer.write_plain_text(
os.path.join(conf_parser.get_out_dir(),
date_prefix + "_" + organism + ".txt"))
if "tsv" in args.output_formats:
logger.info("Writing descriptions to tsv")
desc_writer.write_tsv(
os.path.join(conf_parser.get_out_dir(),
date_prefix + "_" + organism + ".tsv"))
if "ace" in args.output_formats:
logger.info("Writing descriptions to ace")
curators = ["WBPerson324", "WBPerson37462"]
release_version = conf_parser.get_wb_release()
desc_writer.write_ace(
os.path.join(conf_parser.get_out_dir(),
date_prefix + "_" + organism + ".ace"), curators,
release_version)
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)