def score_producer(data, scorer, loader, r_server, lookup_data,
datasources_to_datatypes, dry_run):
target, disease, evidence, is_direct = data
logger = logging.getLogger(__name__)
if evidence:
score = scorer.score(target, disease, evidence, is_direct,
datasources_to_datatypes)
# skip associations only with data with score 0
if score:
gene_data = Gene()
try:
gene_data.load_json(
lookup_data.available_genes.get_gene(target, r_server))
except KeyError as e:
logger.debug('Cannot find gene code "%s" '
'in lookup table' % target)
raise e
score.set_target_data(gene_data)
# create a hpa expression empty jsonserializable class
# to fill from Redis cache lookup_data
hpa_data = HPAExpression()
try:
hpa_data.update(
lookup_data.available_hpa.get_hpa(target, r_server))
except KeyError:
pass
except Exception as e:
raise e
try:
score.set_hpa_data(hpa_data)
except KeyError:
pass
except Exception as e:
raise e
disease_data = EFO()
try:
disease_data.load_json(
lookup_data.available_efos.get_efo(disease, r_server))
except KeyError as e:
logger.debug('Cannot find EFO code "%s" '
'in lookup table' % disease)
logger.exception(e)
score.set_disease_data(disease_data)
element_id = '%s-%s' % (target, disease)
if not dry_run:
loader.put(Const.ELASTICSEARCH_DATA_ASSOCIATION_INDEX_NAME,
Const.ELASTICSEARCH_DATA_ASSOCIATION_DOC_NAME,
element_id, score)
else:
logger.warning('Skipped association with score 0: %s-%s' %
(target, disease))
def merge_data(self, genes, loader, r_server, data_config):
esquery = ESQuery(loader.es)
try:
count = esquery.count_elements_in_index(
Const.ELASTICSEARCH_ENSEMBL_INDEX_NAME)
except NotFoundError as ex:
self._logger.error(
'no Ensembl index in ES. Skipping. Has the --ensembl step been run? Are you pointing to the correct index? %s'
% ex)
raise ex
for row in esquery.get_all_ensembl_genes():
if row['id'] in genes:
gene = genes.get_gene(row['id'])
gene.load_ensembl_data(row)
genes.add_gene(gene)
else:
gene = Gene()
gene.load_ensembl_data(row)
genes.add_gene(gene)
self._clean_non_reference_genes(genes)
self._logger.info("STATS AFTER ENSEMBL PARSING:\n" + genes.get_stats())
def score_producer(data,
scorer, lookup_data, datasources_to_datatypes, dry_run):
target, disease, evidence, is_direct = data
if evidence:
score = scorer.score(target, disease, evidence, is_direct,
datasources_to_datatypes)
# skip associations only with data with score 0
if score:
gene_data = Gene()
gene_data_index = lookup_data.available_genes.get_gene(target)
if gene_data_index != None:
gene_data.load_json(gene_data_index)
score.set_target_data(gene_data)
# create a hpa expression empty jsonserializable class
hpa_data = HPAExpression()
try:
hpa_index = lookup_data.available_hpa.get_hpa(target)
if hpa_index is not None:
hpa_data.update(hpa_index)
except KeyError:
pass
except Exception as e:
raise e
try:
score.set_hpa_data(hpa_data)
except KeyError:
pass
except Exception as e:
raise e
disease_data = EFO()
disease_data.load_json(
lookup_data.available_efos.get_efo(disease))
score.set_disease_data(disease_data)
element_id = '%s-%s' % (target, disease)
#convert the score into a JSON-compatible object
#otherwise Python serialization consumes too much memory
return (element_id, score.to_json())
return None
def merge_data(self, genes, loader, r_server, data_config):
self._logger.info("HGNC parsing - requesting from URL %s",
data_config.hgnc_complete_set)
with URLZSource(data_config.hgnc_complete_set).open() as source:
data = json.load(source)
for row in data['response']['docs']:
gene = Gene()
gene.load_hgnc_data_from_json(row)
genes.add_gene(gene)
self._logger.info("STATS AFTER HGNC PARSING:\n" +
genes.get_stats())
def merge_data(self, genes, es, r_server, data_config, es_config):
index = es_config.ens.name
for row in Search().using(es).index(index).query(MatchAll()).scan():
gene = None
if row['id'] in genes:
gene = genes.get_gene(row['id'])
else:
gene = Gene()
self.load_ensembl_data(gene, row)
genes.add_gene(gene)
self._clean_non_reference_genes(genes)
self._logger.info("STATS AFTER ENSEMBL PARSING:\n" + genes.get_stats())
def merge_data(self, genes, es, r_server, data_config, es_config):
ensembl_filename = data_config.ensembl_filename
with URLZSource(ensembl_filename).open() as ensembl_file:
for line in ensembl_file:
content = json.loads(line)
if content['id'] in genes:
gene = genes.get_gene(content['id'])
else:
gene = Gene()
self.load_ensembl_data(gene, content)
genes.add_gene(gene)
self._clean_non_reference_genes(genes)
self._logger.info("STATS AFTER ENSEMBL PARSING:\n" + genes.get_stats())
def _get_gene_obj(self, geneid):
gene = Gene(geneid)
gene.load_json(self.available_genes.get_gene(geneid))
return gene