def set_sister_species_sentence(dm: WBDataManager,
conf_parser: GenedescConfigParser,
sister_sp_fullname, sister_df: WBDataManager,
species, organism, gene_desc: GeneDescription,
gene: Gene):
best_ortholog = dm.get_best_orthologs_for_gene(
gene_desc.gene_id,
orth_species_full_name=[sister_sp_fullname],
sister_species_data_fetcher=sister_df,
ecode_priority_list=[
"EXP", "IDA", "IPI", "IMP", "IGI", "IEP", "HTP", "HDA", "HMP",
"HGI", "HEP"
])[0][0]
if not best_ortholog[0].startswith("WB:"):
best_ortholog[0] = "WB:" + best_ortholog[0]
sister_sentences_generator = OntologySentenceGenerator(
gene_id=best_ortholog[0],
module=Module.GO,
data_manager=sister_df,
config=conf_parser,
humans=sister_sp_fullname == "H**o sapiens",
limit_to_group="EXPERIMENTAL")
sister_sp_module_sentences = sister_sentences_generator.get_module_sentences(
aspect='P',
qualifier="involved_in",
merge_groups_with_same_prefix=True,
keep_only_best_group=True)
if sister_sp_module_sentences.contains_sentences():
gene_desc.set_or_extend_module_description_and_final_stats(
module=Module.SISTER_SP,
description="in " +
species[species[organism]["main_sister_species"]]["name"] + ", " +
best_ortholog[1] + " " +
sister_sp_module_sentences.get_description())
def test_expression_the_cell_renaming_to_widely(self):
self.df.load_ontology_from_file(
ontology_type=DataType.EXPR,
ontology_url=self.df.expression_ontology_url,
ontology_cache_path=self.df.expression_ontology_cache_path,
config=self.conf_parser)
self.df.load_associations_from_file(
associations_type=DataType.EXPR,
associations_url=self.df.expression_associations_url,
associations_cache_path=self.df.expression_associations_cache_path,
config=self.conf_parser)
gene_desc = GeneDescription(gene_id="WB:WBGene00007352",
gene_name="cdc-48.1",
add_gene_name=False)
expr_sentence_generator = OntologySentenceGenerator(
gene_id=gene_desc.gene_id,
module=Module.EXPRESSION,
data_manager=self.df,
config=self.conf_parser)
expression_module_sentences = expr_sentence_generator.get_module_sentences(
config=self.conf_parser,
aspect='A',
qualifier="Verified",
merge_groups_with_same_prefix=True,
keep_only_best_group=False)
gene_desc.set_or_extend_module_description_and_final_stats(
module_sentences=expression_module_sentences,
module=Module.EXPRESSION)
self.assertTrue("is expressed widely" in gene_desc.description)
def set_information_poor_sentence(orth_fullnames: List[str], selected_orthologs, ensembl_hgnc_ids_map,
conf_parser: GenedescConfigParser, human_df_agr: DataManager,
gene_desc: GeneDescription, dm: WBDataManager, gene: Gene):
if len(orth_fullnames) == 1 and orth_fullnames[0] == "H**o sapiens":
best_orth = get_best_human_ortholog_for_info_poor(selected_orthologs, ensembl_hgnc_ids_map,
conf_parser.get_annotations_priority(module=Module.GO),
human_df_agr, config=conf_parser)
if best_orth:
if not best_orth.startswith("RGD:"):
best_orth = "RGD:" + best_orth
human_go_sent_generator = OntologySentenceGenerator(gene_id=best_orth, module=Module.GO,
data_manager=human_df_agr, config=conf_parser,
humans=False, limit_to_group="EXPERIMENTAL")
human_func_module_sentences = human_go_sent_generator.get_module_sentences(
config=conf_parser, aspect='F', merge_groups_with_same_prefix=True, keep_only_best_group=True)
human_func_sent = human_func_module_sentences.get_description()
if human_func_sent:
gene_desc.set_or_extend_module_description_and_final_stats(
module=Module.INFO_POOR_HUMAN_FUNCTION, description="human " +
human_df_agr.go_associations.subject_label_map[
best_orth] + " " + human_func_sent)
protein_domains = dm.protein_domains[gene_desc.gene_id[3:]]
if protein_domains:
dom_word = "domain"
if len(protein_domains) > 1:
dom_word = "domains"
gene_desc.set_or_extend_module_description_and_final_stats(
module=Module.PROTEIN_DOMAIN,
description="is predicted to encode a protein with the following " + dom_word + ": " +
concatenate_words_with_oxford_comma([ptdom[1] if ptdom[1] != "" else ptdom[0] for
ptdom in protein_domains]))
def set_expression_module(df: DataManager, conf_parser: GenedescConfigParser, gene_desc: GeneDescription, gene: Gene):
expr_sentence_generator = OntologySentenceGenerator(gene_id=gene.id, module=Module.EXPRESSION, data_manager=df,
config=conf_parser)
expression_module_sentences = expr_sentence_generator.get_module_sentences(
aspect='A', qualifier="Verified", merge_groups_with_same_prefix=True, keep_only_best_group=False)
gene_desc.set_or_extend_module_description_and_final_stats(module_sentences=expression_module_sentences,
module=Module.EXPRESSION)
gene_desc.set_or_update_initial_stats(module=Module.EXPRESSION, sent_generator=expr_sentence_generator,
module_sentences=expression_module_sentences)
def set_orthology_sentence(dm: WBDataManager, orth_fullnames: List[str], gene_desc: GeneDescription,
human_genes_props, api_manager):
best_orthologs, selected_orth_name = dm.get_best_orthologs_for_gene(gene_desc.gene_id,
orth_species_full_name=orth_fullnames)
selected_orthologs = []
if best_orthologs:
gene_desc.stats.set_best_orthologs = [orth[0] for orth in best_orthologs]
if len(orth_fullnames) == 1 and orth_fullnames[0] == "H**o sapiens":
sel_orthologs, orth_sent = generate_ortholog_sentence_wormbase_human(best_orthologs, human_genes_props)
selected_orthologs = [orth for orth in best_orthologs if orth[1] in sel_orthologs]
else:
orth_sent = generate_ortholog_sentence_wormbase_non_c_elegans(best_orthologs, selected_orth_name,
api_manager=api_manager)
gene_desc.set_or_extend_module_description_and_final_stats(module=Module.ORTHOLOGY, description=orth_sent)
return selected_orthologs
def set_alliance_human_orthology_module(orthologs: List[List[str]], gene_desc: GeneDescription,
config: GenedescConfigParser, excluded_orthologs: bool = False):
"""set orthology module for Alliance human orthologs
Args:
orthologs (List[List[str]]): list of human orthologs, containing gene_id, gene_symbol, and gene_name
gene_desc (GeneDescription): the gene description object to update
config (GenedescConfigParser): a gene descriptions configuration object
excluded_orthologs (bool): whether some of the orthologs have been excluded from the final set. If true, the
final sentence will include a prefix to specify that some orthologs have been omitted
"""
if len(orthologs) > 0:
prefix = "human"
orthologs_display = sorted(orthologs, key=lambda x: x[2])
if excluded_orthologs or len(orthologs) > 3:
orthologs_display = orthologs_display[0:3]
prefix = "several human genes including"
sentence = "orthologous to " + prefix + " " + concatenate_words_with_oxford_comma(
[orth[1] + " (" + orth[2] + ")" if orth[2] else orth[1] for orth in orthologs_display],
separator=config.get_terms_delimiter())
gene_desc.set_or_extend_module_description_and_final_stats(module=Module.ORTHOLOGY, description=sentence)
def get_generators(self, data_provider, gd_data_manager, gd_config,
json_desc_writer):
"""Create generators."""
gene_prefix = ""
if data_provider == "HUMAN":
return_set = Neo4jHelper.run_single_parameter_query(
self.get_all_genes_human_query, "RGD")
gene_prefix = "RGD:"
else:
return_set = Neo4jHelper.run_single_parameter_query(
self.get_all_genes_query, data_provider)
descriptions = []
best_orthologs = self.get_best_orthologs_from_db(
data_provider=data_provider)
for record in return_set:
gene = Gene(id=gene_prefix + record["g.primaryKey"],
name=record["g.symbol"],
dead=False,
pseudo=False)
gene_desc = GeneDescription(gene_id=record["g.primaryKey"],
gene_name=gene.name,
add_gene_name=False,
config=gd_config)
set_gene_ontology_module(dm=gd_data_manager,
conf_parser=gd_config,
gene_desc=gene_desc,
gene=gene)
set_expression_module(df=gd_data_manager,
conf_parser=gd_config,
gene_desc=gene_desc,
gene=gene)
set_disease_module(df=gd_data_manager,
conf_parser=gd_config,
gene_desc=gene_desc,
gene=gene,
human=data_provider == "HUMAN")
if gene.id in best_orthologs:
gene_desc.stats.set_best_orthologs = best_orthologs[gene.id][0]
set_alliance_human_orthology_module(
orthologs=best_orthologs[gene.id][0],
excluded_orthologs=best_orthologs[gene.id][1],
gene_desc=gene_desc,
config=gd_config)
if gene_desc.description:
descriptions.append({
"genePrimaryKey": gene_desc.gene_id,
"geneDescription": gene_desc.description
})
json_desc_writer.add_gene_desc(gene_desc)
yield [descriptions]
def test_set_or_extend_module_description_and_final_stats(self):
gene_desc = GeneDescription(gene_id="FB:FBgn0027655", gene_name="Test gene", add_gene_name=False)
go_sent_generator = OntologySentenceGenerator(gene_id="FB:FBgn0027655", module=Module.GO,
data_manager=self.df, config=self.conf_parser)
sentences = go_sent_generator.get_module_sentences(config=self.conf_parser, aspect='P',
qualifier='', merge_groups_with_same_prefix=True,
keep_only_best_group=True)
gene_desc.set_or_extend_module_description_and_final_stats(module=Module.GO_PROCESS, module_sentences=sentences)
self.assertTrue(gene_desc.description, "Is involved in several processes, including axo-dendritic transport, "
"establishment of mitotic spindle orientation, and positive regulation "
"of extent of heterochromatin assembly")
gene_desc = GeneDescription(gene_id="FB:FBgn0027655", gene_name="Test gene", add_gene_name=True)
gene_desc.set_or_extend_module_description_and_final_stats(module=Module.GO_PROCESS, module_sentences=sentences)
self.assertTrue(gene_desc.description, "Test gene is involved in several processes, including axo-dendritic "
"transport, establishment of mitotic spindle orientation, and positive "
"regulation of extent of heterochromatin assembly")
def test_trimming_lca(self):
self.conf_parser.config["go_sentences_options"]["trimming_algorithm"] = "ic"
self.conf_parser.config["expression_sentences_options"]["trimming_algorithm"] = "ic"
gene = Gene(id="WB:WBGene00000018", name="abl-1", dead=False, pseudo=False)
self.df.load_ontology_from_file(ontology_type=DataType.EXPR, ontology_url="file://" + os.path.join(
self.this_dir, "data", "anatomy_ontology.WS274.obo"),
ontology_cache_path=os.path.join(self.this_dir, "cache",
"anatomy_ontology.WS274.obo"),
config=self.conf_parser)
logger.info("Loading expression associations from file")
self.conf_parser.config["expression_sentences_options"]["max_num_terms"] = 5
self.conf_parser.config["expression_sentences_options"]["trim_min_distance_from_root"]["A"] = 4
self.conf_parser.config["expression_sentences_options"]["remove_children_if_parent_is_present"] = False
associations = self.get_associations(gene.id, ["WBbt:0006796", "WBbt:0006759", "WBbt:0005300", "WBbt:0008598",
"WBbt:0003681", "WBbt:0005829", "WBbt:0003927", "WBbt:0006751"],
["Verified"], "A", "IDA")
self.df.expression_associations = AssociationSetFactory().create_from_assocs(assocs=associations,
ontology=self.df.expression_ontology)
self.conf_parser.config["go_sentences_options"]["trimming_algorithm"] = "lca"
self.conf_parser.config["expression_sentences_options"]["trimming_algorithm"] = "lca"
gene_desc_lca = GeneDescription(gene_id=gene.id, config=self.conf_parser, gene_name="abl-1",
add_gene_name=False)
set_gene_ontology_module(dm=self.df, conf_parser=self.conf_parser, gene_desc=gene_desc_lca, gene=gene)
set_expression_module(self.df, self.conf_parser, gene_desc_lca, gene)
gene_desc_lca.stats.calculate_stats(data_manager=self.df)
self.conf_parser.config["go_sentences_options"]["trimming_algorithm"] = "ic"
self.conf_parser.config["expression_sentences_options"]["trimming_algorithm"] = "ic"
set_ic_ontology_struct(ontology=self.df.go_ontology, relations=self.df.go_relations)
set_ic_ontology_struct(ontology=self.df.expression_ontology, relations=self.df.expr_relations)
gene_desc_ic = GeneDescription(gene_id=gene.id, config=self.conf_parser, gene_name="abl-1",
add_gene_name=False)
set_gene_ontology_module(dm=self.df, conf_parser=self.conf_parser, gene_desc=gene_desc_ic, gene=gene)
set_expression_module(self.df, self.conf_parser, gene_desc_ic, gene)
gene_desc_ic.stats.calculate_stats(data_manager=self.df)
self.assertTrue(gene_desc_lca.stats.coverage_percentage >= gene_desc_ic.stats.coverage_percentage, "1")
self.conf_parser.config["go_sentences_options"]["trimming_algorithm"] = "lca"
self.conf_parser.config["expression_sentences_options"]["trimming_algorithm"] = "lca"
gene = Gene(id="WB:WBGene00000022", name="aat-1", dead=False, pseudo=False)
associations = self.get_associations(gene.id, ["WBbt:0005828", "WBbt:0006751", "WBbt:0005439", "WBbt:0005788",
"WBbt:0006749", "WBbt:0005300", "WBbt:0005735", "WBbt:0005747",
"WBbt:0005772", "WBbt:0005776", "WBbt:0005812", "WBbt:0005741",
"WBbt:0005799", "WBbt:0003681"],
["Verified"], "A", "IDA")
self.df.expression_associations = AssociationSetFactory().create_from_assocs(
assocs=associations, ontology=self.df.expression_ontology)
gene_desc_lca = GeneDescription(gene_id=gene.id, config=self.conf_parser, gene_name="aat-1",
add_gene_name=False)
set_gene_ontology_module(dm=self.df, conf_parser=self.conf_parser, gene_desc=gene_desc_lca, gene=gene)
set_expression_module(self.df, self.conf_parser, gene_desc_lca, gene)
gene_desc_lca.stats.calculate_stats(data_manager=self.df)
self.conf_parser.config["go_sentences_options"]["trimming_algorithm"] = "ic"
self.conf_parser.config["expression_sentences_options"]["trimming_algorithm"] = "ic"
gene_desc_ic = GeneDescription(gene_id=gene.id, config=self.conf_parser, gene_name="aat-1",
add_gene_name=False)
set_gene_ontology_module(dm=self.df, conf_parser=self.conf_parser, gene_desc=gene_desc_ic, gene=gene)
set_expression_module(self.df, self.conf_parser, gene_desc_ic, gene)
gene_desc_ic.stats.calculate_stats(data_manager=self.df)
self.assertTrue(gene_desc_lca.stats.coverage_percentage >= gene_desc_ic.stats.coverage_percentage, "2")
self.conf_parser.config["go_sentences_options"]["trimming_algorithm"] = "lca"
self.conf_parser.config["expression_sentences_options"]["trimming_algorithm"] = "lca"
gene = Gene(id="WB:WBGene00000044", name="acr-5", dead=False, pseudo=False)
associations = self.get_associations(gene.id, ['WBbt:0003679', 'WBbt:0006759', 'WBbt:0005336', 'WBbt:0006751',
'WBbt:0005300', 'WBbt:0005274', 'WBbt:0005741', 'WBbt:0006749',
'WBbt:0005735'],
["Verified"], "A", "IDA")
self.df.expression_associations = AssociationSetFactory().create_from_assocs(
assocs=associations, ontology=self.df.expression_ontology)
gene_desc_lca = GeneDescription(gene_id=gene.id, config=self.conf_parser, gene_name="acr-5",
add_gene_name=False)
set_gene_ontology_module(dm=self.df, conf_parser=self.conf_parser, gene_desc=gene_desc_lca, gene=gene)
set_expression_module(self.df, self.conf_parser, gene_desc_lca, gene)
gene_desc_lca.stats.calculate_stats(data_manager=self.df)
self.conf_parser.config["go_sentences_options"]["trimming_algorithm"] = "ic"
self.conf_parser.config["expression_sentences_options"]["trimming_algorithm"] = "ic"
gene_desc_ic = GeneDescription(gene_id=gene.id, config=self.conf_parser, gene_name="acr-5",
add_gene_name=False)
set_gene_ontology_module(dm=self.df, conf_parser=self.conf_parser, gene_desc=gene_desc_ic, gene=gene)
set_expression_module(self.df, self.conf_parser, gene_desc_ic, gene)
gene_desc_ic.stats.calculate_stats(data_manager=self.df)
self.assertTrue(gene_desc_lca.stats.coverage_percentage >= gene_desc_ic.stats.coverage_percentage, "3")
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 set_expression_cluster_sentence(dm: WBDataManager,
conf_parser: GenedescConfigParser,
gene_desc: GeneDescription, gene: Gene,
api_manager: APIManager):
expr_sentence_generator = OntologySentenceGenerator(
gene_id=gene.id,
module=Module.EXPRESSION,
data_manager=dm,
config=conf_parser)
ec_gene_id = gene_desc.gene_id[3:]
ec_anatomy_studies = dm.get_expression_cluster_feature(
gene_id=ec_gene_id,
expression_cluster_type=ExpressionClusterType.ANATOMY,
feature=ExpressionClusterFeature.STUDIES)
ec_anatomy_terms = dm.get_expression_cluster_feature(
gene_id=ec_gene_id,
feature=ExpressionClusterFeature.TERMS,
expression_cluster_type=ExpressionClusterType.ANATOMY)
if dm.expression_ontology is not None:
expression_enriched_module_sentences = expr_sentence_generator.get_module_sentences(
aspect='A',
qualifier="Enriched",
merge_groups_with_same_prefix=True,
keep_only_best_group=False)
gene_desc.set_or_extend_module_description_and_final_stats(
module=Module.EXPRESSION_CLUSTER_ANATOMY,
description=expression_enriched_module_sentences.get_description(),
additional_postfix_terms_list=ec_anatomy_studies,
additional_postfix_final_word="studies",
use_single_form=True)
elif ec_anatomy_terms:
gene_desc.set_or_extend_module_description_and_final_stats(
module=Module.EXPRESSION_CLUSTER_ANATOMY,
description="is enriched in " +
concatenate_words_with_oxford_comma(
ec_anatomy_terms,
separator=conf_parser.get_terms_delimiter()) + " based on",
additional_postfix_terms_list=ec_anatomy_studies,
additional_postfix_final_word="studies",
use_single_form=True)
ec_molreg_terms = dm.get_expression_cluster_feature(
gene_id=ec_gene_id,
expression_cluster_type=ExpressionClusterType.MOLREG,
feature=ExpressionClusterFeature.TERMS)
ec_molreg_studies = dm.get_expression_cluster_feature(
gene_id=ec_gene_id,
feature=ExpressionClusterFeature.STUDIES,
expression_cluster_type=ExpressionClusterType.MOLREG)
ec_genereg_terms = dm.get_expression_cluster_feature(
gene_id=ec_gene_id,
expression_cluster_type=ExpressionClusterType.GENEREG,
feature=ExpressionClusterFeature.TERMS)
ec_genereg_studies = dm.get_expression_cluster_feature(
gene_id=ec_gene_id,
feature=ExpressionClusterFeature.STUDIES,
expression_cluster_type=ExpressionClusterType.GENEREG)
if ec_genereg_terms:
several_word = ""
if len(ec_genereg_terms) > 3:
t_p = [
t_p for t_p in sorted(
[[term, api_manager.get_textpresso_popularity(term)]
for term in ec_genereg_terms],
key=lambda x: (x[1], x[0][1]),
reverse=True)
]
ec_genereg_terms = [term for term, popularity in t_p[0:3]]
several_word = "several genes including "
gene_desc.set_or_extend_module_description_and_final_stats(
module=Module.EXPRESSION_CLUSTER_GENE,
description="is affected by " + several_word +
concatenate_words_with_oxford_comma(
ec_genereg_terms,
separator=conf_parser.get_terms_delimiter()) + " based on",
additional_postfix_terms_list=ec_genereg_studies,
additional_postfix_final_word="studies",
use_single_form=True)
if ec_molreg_terms:
several_word = ""
if len(ec_molreg_terms) > 3:
several_word = num2words(
len(ec_molreg_terms)) + " chemicals including "
gene_desc.set_or_extend_module_description_and_final_stats(
module=Module.EXPRESSION_CLUSTER_MOLECULE,
description="is affected by " + several_word +
concatenate_words_with_oxford_comma(
ec_molreg_terms[0:3],
separator=conf_parser.get_terms_delimiter()) + " based on",
additional_postfix_terms_list=ec_molreg_studies,
additional_postfix_final_word="studies",
use_single_form=True)
def set_disease_module(df: DataManager, conf_parser: GenedescConfigParser, gene_desc: GeneDescription, gene: Gene,
human: bool = False):
do_sentence_exp_generator = OntologySentenceGenerator(gene_id=gene.id,
module=Module.DO_EXPERIMENTAL, data_manager=df,
config=conf_parser, limit_to_group="EXPERIMENTAL",
humans=human)
disease_exp_module_sentences = do_sentence_exp_generator.get_module_sentences(
aspect='D', merge_groups_with_same_prefix=True, keep_only_best_group=False)
gene_desc.set_or_extend_module_description_and_final_stats(module=Module.DO_EXPERIMENTAL,
module_sentences=disease_exp_module_sentences)
do_sentence_bio_generator = OntologySentenceGenerator(gene_id=gene.id, module=Module.DO_BIOMARKER,
data_manager=df, config=conf_parser,
limit_to_group="BIOMARKER", humans=human)
disease_bio_module_sentences = do_sentence_bio_generator.get_module_sentences(
aspect='D', merge_groups_with_same_prefix=True, keep_only_best_group=False)
gene_desc.set_or_extend_module_description_and_final_stats(module=Module.DO_BIOMARKER,
module_sentences=disease_bio_module_sentences)
do_via_orth_sentence_generator = OntologySentenceGenerator(
gene_id=gene.id, module=Module.DO_ORTHOLOGY, data_manager=df, config=conf_parser, humans=human)
disease_via_orth_module_sentences = do_via_orth_sentence_generator.get_module_sentences(
aspect='D', merge_groups_with_same_prefix=True, keep_only_best_group=False)
gene_desc.set_or_extend_module_description_and_final_stats(module=Module.DO_ORTHOLOGY,
module_sentences=disease_via_orth_module_sentences)
gene_desc.set_or_update_initial_stats(module=Module.DO_EXPERIMENTAL, sent_generator=do_sentence_exp_generator,
module_sentences=disease_exp_module_sentences)
gene_desc.set_or_update_initial_stats(module=Module.DO_BIOMARKER, sent_generator=do_sentence_bio_generator,
module_sentences=disease_bio_module_sentences)
gene_desc.set_or_update_initial_stats(module=Module.DO_ORTHOLOGY, sent_generator=do_via_orth_sentence_generator,
module_sentences=disease_via_orth_module_sentences)
def set_gene_ontology_module(dm: DataManager, conf_parser: GenedescConfigParser, gene_desc: GeneDescription,
gene: Gene):
go_sent_generator_exp = OntologySentenceGenerator(gene_id=gene.id, module=Module.GO, data_manager=dm,
config=conf_parser, limit_to_group="EXPERIMENTAL")
go_sent_generator = OntologySentenceGenerator(gene_id=gene.id, module=Module.GO, data_manager=dm,
config=conf_parser)
# Generate sentences with experimental annotations only
func_module_sentences_contributes_to = go_sent_generator_exp.get_module_sentences(
aspect='F', qualifier='contributes_to', merge_groups_with_same_prefix=True, keep_only_best_group=True)
func_module_sentences_noq = go_sent_generator_exp.get_module_sentences(
aspect='F', qualifier='', merge_groups_with_same_prefix=True, keep_only_best_group=True)
func_module_sentences_enables = go_sent_generator_exp.get_module_sentences(
aspect='F', qualifier='enables', merge_groups_with_same_prefix=True, keep_only_best_group=True)
# If experimental sentences are all empty, generate sentences with all annotations
if not func_module_sentences_contributes_to.contains_sentences() and \
not func_module_sentences_noq.contains_sentences() and \
not func_module_sentences_enables.contains_sentences():
func_module_sentences_contributes_to = go_sent_generator.get_module_sentences(
aspect='F', qualifier='contributes_to', merge_groups_with_same_prefix=True, keep_only_best_group=True)
func_module_sentences_noq = go_sent_generator.get_module_sentences(
aspect='F', qualifier='', merge_groups_with_same_prefix=True, keep_only_best_group=True)
func_module_sentences_enables = go_sent_generator.get_module_sentences(
aspect='F', qualifier='enables', merge_groups_with_same_prefix=True, keep_only_best_group=True)
gene_desc.set_or_extend_module_description_and_final_stats(module_sentences=func_module_sentences_noq,
module=Module.GO_FUNCTION)
gene_desc.set_or_extend_module_description_and_final_stats(module_sentences=func_module_sentences_enables,
module=Module.GO_FUNCTION)
gene_desc.set_or_extend_module_description_and_final_stats(module_sentences=func_module_sentences_contributes_to,
module=Module.GO_FUNCTION)
proc_module_sentences_noq = go_sent_generator_exp.get_module_sentences(
aspect='P', qualifier='', merge_groups_with_same_prefix=True, keep_only_best_group=True)
proc_module_sentences_inv = go_sent_generator_exp.get_module_sentences(
aspect='P', qualifier='involved_in', merge_groups_with_same_prefix=True, keep_only_best_group=True)
proc_module_sentences_acts_pos = go_sent_generator_exp.get_module_sentences(
aspect='P', qualifier='acts_upstream_of_positive_effect', merge_groups_with_same_prefix=True,
keep_only_best_group=True)
proc_module_sentences_acts_neg = go_sent_generator_exp.get_module_sentences(
aspect='P', qualifier='acts_upstream_of_negative_effect', merge_groups_with_same_prefix=True,
keep_only_best_group=True)
proc_module_sentences_acts_with_pos = go_sent_generator_exp.get_module_sentences(
aspect='P', qualifier='acts_upstream_of_or_within_positive_effect', merge_groups_with_same_prefix=True,
keep_only_best_group=True)
proc_module_sentences_acts_with_neg = go_sent_generator_exp.get_module_sentences(
aspect='P', qualifier='acts_upstream_of_or_within_negative_effect', merge_groups_with_same_prefix=True,
keep_only_best_group=True)
proc_module_sentences_acts_with = go_sent_generator_exp.get_module_sentences(
aspect='P', qualifier='acts_upstream_of_or_within', merge_groups_with_same_prefix=True,
keep_only_best_group=True)
if not proc_module_sentences_noq.contains_sentences() and \
not proc_module_sentences_inv.contains_sentences() and \
not proc_module_sentences_acts_pos.contains_sentences() and \
not proc_module_sentences_acts_neg.contains_sentences() and \
not proc_module_sentences_acts_with_pos.contains_sentences() and \
not proc_module_sentences_acts_with_neg.contains_sentences() and \
not proc_module_sentences_acts_with.contains_sentences():
proc_module_sentences_noq = go_sent_generator.get_module_sentences(
aspect='P', qualifier='', merge_groups_with_same_prefix=True, keep_only_best_group=True)
proc_module_sentences_inv = go_sent_generator.get_module_sentences(
aspect='P', qualifier='involved_in', merge_groups_with_same_prefix=True, keep_only_best_group=True)
proc_module_sentences_acts_pos = go_sent_generator.get_module_sentences(
aspect='P', qualifier='acts_upstream_of_positive_effect', merge_groups_with_same_prefix=True,
keep_only_best_group=True)
proc_module_sentences_acts_neg = go_sent_generator.get_module_sentences(
aspect='P', qualifier='acts_upstream_of_negative_effect', merge_groups_with_same_prefix=True,
keep_only_best_group=True)
proc_module_sentences_acts_with_pos = go_sent_generator.get_module_sentences(
aspect='P', qualifier='acts_upstream_of_or_within_positive_effect', merge_groups_with_same_prefix=True,
keep_only_best_group=True)
proc_module_sentences_acts_with_neg = go_sent_generator.get_module_sentences(
aspect='P', qualifier='acts_upstream_of_or_within_negative_effect', merge_groups_with_same_prefix=True,
keep_only_best_group=True)
proc_module_sentences_acts_with = go_sent_generator.get_module_sentences(
aspect='P', qualifier='acts_upstream_of_or_within', merge_groups_with_same_prefix=True,
keep_only_best_group=True)
gene_desc.set_or_extend_module_description_and_final_stats(
module_sentences=proc_module_sentences_noq, module=Module.GO_PROCESS)
gene_desc.set_or_extend_module_description_and_final_stats(
module_sentences=proc_module_sentences_inv, module=Module.GO_PROCESS)
gene_desc.set_or_extend_module_description_and_final_stats(
module_sentences=proc_module_sentences_acts_pos, module=Module.GO_PROCESS)
gene_desc.set_or_extend_module_description_and_final_stats(
module_sentences=proc_module_sentences_acts_neg, module=Module.GO_PROCESS)
gene_desc.set_or_extend_module_description_and_final_stats(
module_sentences=proc_module_sentences_acts_with_pos, module=Module.GO_PROCESS)
gene_desc.set_or_extend_module_description_and_final_stats(
module_sentences=proc_module_sentences_acts_with_neg, module=Module.GO_PROCESS)
gene_desc.set_or_extend_module_description_and_final_stats(
module_sentences=proc_module_sentences_acts_with, module=Module.GO_PROCESS)
comp_module_sentence_colocalizes_with = go_sent_generator_exp.get_module_sentences(
aspect='C', qualifier='colocalizes_with', merge_groups_with_same_prefix=True,
keep_only_best_group=True)
comp_module_sentences_noq = go_sent_generator_exp.get_module_sentences(
aspect='C', qualifier='', merge_groups_with_same_prefix=True, keep_only_best_group=True)
comp_module_sentences_located = go_sent_generator_exp.get_module_sentences(
aspect='C', qualifier='located_in', merge_groups_with_same_prefix=True, keep_only_best_group=True)
comp_module_sentences_part = go_sent_generator_exp.get_module_sentences(
aspect='C', qualifier='part_of', merge_groups_with_same_prefix=True, keep_only_best_group=True)
comp_module_sentences_active = go_sent_generator_exp.get_module_sentences(
aspect='C', qualifier='is_active_in', merge_groups_with_same_prefix=True, keep_only_best_group=True)
if not comp_module_sentence_colocalizes_with.contains_sentences() and \
not comp_module_sentences_noq.contains_sentences() and \
not comp_module_sentences_located.contains_sentences() and \
not comp_module_sentences_part.contains_sentences() and \
not comp_module_sentences_active.contains_sentences():
comp_module_sentence_colocalizes_with = go_sent_generator.get_module_sentences(
aspect='C', qualifier='colocalizes_with', merge_groups_with_same_prefix=True,
keep_only_best_group=True)
comp_module_sentences_noq = go_sent_generator.get_module_sentences(
aspect='C', qualifier='', merge_groups_with_same_prefix=True, keep_only_best_group=True)
comp_module_sentences_located = go_sent_generator.get_module_sentences(
aspect='C', qualifier='located_in', merge_groups_with_same_prefix=True, keep_only_best_group=True)
comp_module_sentences_part = go_sent_generator.get_module_sentences(
aspect='C', qualifier='part_of', merge_groups_with_same_prefix=True, keep_only_best_group=True)
comp_module_sentences_active = go_sent_generator.get_module_sentences(
aspect='C', qualifier='is_active_in', merge_groups_with_same_prefix=True, keep_only_best_group=True)
gene_desc.set_or_extend_module_description_and_final_stats(
module_sentences=comp_module_sentences_noq, module=Module.GO_COMPONENT)
gene_desc.set_or_extend_module_description_and_final_stats(
module_sentences=comp_module_sentences_located, module=Module.GO_COMPONENT)
gene_desc.set_or_extend_module_description_and_final_stats(
module_sentences=comp_module_sentences_part, module=Module.GO_COMPONENT)
gene_desc.set_or_extend_module_description_and_final_stats(
module_sentences=comp_module_sentences_active, module=Module.GO_COMPONENT)
gene_desc.set_or_extend_module_description_and_final_stats(module_sentences=comp_module_sentence_colocalizes_with,
module=Module.GO_COMPONENT)
gene_desc.set_or_update_initial_stats(module=Module.GO_FUNCTION, sent_generator=go_sent_generator,
module_sentences=func_module_sentences_contributes_to)
gene_desc.set_or_update_initial_stats(module=Module.GO_FUNCTION, sent_generator=go_sent_generator,
module_sentences=func_module_sentences_noq)
gene_desc.set_or_update_initial_stats(module=Module.GO_FUNCTION, sent_generator=go_sent_generator,
module_sentences=func_module_sentences_enables)
gene_desc.set_or_update_initial_stats(module=Module.GO_PROCESS, sent_generator=go_sent_generator,
module_sentences=proc_module_sentences_noq)
gene_desc.set_or_update_initial_stats(module=Module.GO_PROCESS, sent_generator=go_sent_generator,
module_sentences=proc_module_sentences_inv)
gene_desc.set_or_update_initial_stats(module=Module.GO_PROCESS, sent_generator=go_sent_generator,
module_sentences=proc_module_sentences_acts_pos)
gene_desc.set_or_update_initial_stats(module=Module.GO_PROCESS, sent_generator=go_sent_generator,
module_sentences=proc_module_sentences_acts_neg)
gene_desc.set_or_update_initial_stats(module=Module.GO_PROCESS, sent_generator=go_sent_generator,
module_sentences=proc_module_sentences_acts_with_pos)
gene_desc.set_or_update_initial_stats(module=Module.GO_PROCESS, sent_generator=go_sent_generator,
module_sentences=proc_module_sentences_acts_with_neg)
gene_desc.set_or_update_initial_stats(module=Module.GO_PROCESS, sent_generator=go_sent_generator,
module_sentences=proc_module_sentences_acts_with)
gene_desc.set_or_update_initial_stats(module=Module.GO_COMPONENT, sent_generator=go_sent_generator,
module_sentences=comp_module_sentence_colocalizes_with)
gene_desc.set_or_update_initial_stats(module=Module.GO_COMPONENT, sent_generator=go_sent_generator,
module_sentences=comp_module_sentences_noq)
gene_desc.set_or_update_initial_stats(module=Module.GO_COMPONENT, sent_generator=go_sent_generator,
module_sentences=comp_module_sentences_located)
gene_desc.set_or_update_initial_stats(module=Module.GO_COMPONENT, sent_generator=go_sent_generator,
module_sentences=comp_module_sentences_part)
gene_desc.set_or_update_initial_stats(module=Module.GO_COMPONENT, sent_generator=go_sent_generator,
module_sentences=comp_module_sentences_active)
def set_gene_ontology_module(dm: DataManager,
conf_parser: GenedescConfigParser,
gene_desc: GeneDescription, gene: Gene):
go_sent_generator_exp = OntologySentenceGenerator(
gene_id=gene.id,
module=Module.GO,
data_manager=dm,
config=conf_parser,
limit_to_group="EXPERIMENTAL")
go_sent_generator = OntologySentenceGenerator(gene_id=gene.id,
module=Module.GO,
data_manager=dm,
config=conf_parser)
contributes_to_module_sentences = go_sent_generator.get_module_sentences(
config=conf_parser,
aspect='F',
qualifier='contributes_to',
merge_groups_with_same_prefix=True,
keep_only_best_group=True)
if contributes_to_module_sentences.contains_sentences():
func_module_sentences = go_sent_generator_exp.get_module_sentences(
config=conf_parser,
aspect='F',
merge_groups_with_same_prefix=True,
keep_only_best_group=True)
gene_desc.set_or_extend_module_description_and_final_stats(
module_sentences=func_module_sentences, module=Module.GO_FUNCTION)
else:
func_module_sentences = go_sent_generator.get_module_sentences(
config=conf_parser,
aspect='F',
merge_groups_with_same_prefix=True,
keep_only_best_group=True)
gene_desc.set_or_extend_module_description_and_final_stats(
module_sentences=func_module_sentences, module=Module.GO_FUNCTION)
gene_desc.set_or_extend_module_description_and_final_stats(
module_sentences=contributes_to_module_sentences,
module=Module.GO_FUNCTION)
proc_module_sentences = go_sent_generator.get_module_sentences(
config=conf_parser,
aspect='P',
merge_groups_with_same_prefix=True,
keep_only_best_group=True)
gene_desc.set_or_extend_module_description_and_final_stats(
module_sentences=proc_module_sentences, module=Module.GO_PROCESS)
colocalizes_with_module_sentences = go_sent_generator.get_module_sentences(
config=conf_parser,
aspect='C',
qualifier='colocalizes_with',
merge_groups_with_same_prefix=True,
keep_only_best_group=True)
if colocalizes_with_module_sentences.contains_sentences():
comp_module_sentences = go_sent_generator_exp.get_module_sentences(
config=conf_parser,
aspect='C',
merge_groups_with_same_prefix=True,
keep_only_best_group=True)
gene_desc.set_or_extend_module_description_and_final_stats(
module_sentences=comp_module_sentences, module=Module.GO_COMPONENT)
else:
comp_module_sentences = go_sent_generator.get_module_sentences(
config=conf_parser,
aspect='C',
merge_groups_with_same_prefix=True,
keep_only_best_group=True)
gene_desc.set_or_extend_module_description_and_final_stats(
module_sentences=comp_module_sentences, module=Module.GO_COMPONENT)
gene_desc.set_or_extend_module_description_and_final_stats(
module_sentences=colocalizes_with_module_sentences,
module=Module.GO_COMPONENT)
gene_desc.set_initial_stats(
module=Module.GO_FUNCTION,
sentence_generator=go_sent_generator,
sentence_generator_exp_only=go_sent_generator_exp)
gene_desc.set_initial_stats(
module=Module.GO_PROCESS,
sentence_generator=go_sent_generator,
sentence_generator_exp_only=go_sent_generator_exp)
gene_desc.set_initial_stats(
module=Module.GO_COMPONENT,
sentence_generator=go_sent_generator,
sentence_generator_exp_only=go_sent_generator_exp)