def get_gene(self, gene_name):
"""
Get a gene from the cache or attempt to disambiguate or add a new
record.
"""
if not gene_name:
return
gene_pk = self.gene_cache.get(gene_name, None)
if gene_pk:
return Gene(pk=gene_pk)
# Attempt to disambiguate, only if this is the only synonym may it be
# associated.
potential_genes = list(Gene.objects.filter(
synonyms__label__iexact=gene_name).distinct())
if len(potential_genes) == 1:
self.gene_cache[gene_name] = potential_genes[0].pk
return potential_genes[0]
# Only if there are no matches should we create a new record,
# otherwise the synonym war will continue
if len(potential_genes) == 0:
gene = Gene(chr=self.get_chromosome(self.chrom), symbol=gene_name)
gene.save()
self.gene_cache[gene_name] = gene.pk
return gene
def setUp(self):
import string
chr1 = Chromosome(value='1', label='1')
chr1.save()
genes = {}
for char in string.lowercase:
g = Gene(chr=chr1, symbol=char)
g.save()
genes[char] = g
geneset = GeneSet(name='test')
geneset.save()
geneset.bulk([genes[c] for c in 'someday'])
self.genes = genes
self.geneset = geneset
def setUp(self):
import string
chr1 = Chromosome(value='1', label='1')
chr1.save()
genes = {}
for char in string.lowercase:
g = Gene(chr=chr1, symbol=char)
g.save()
genes[char] = g
geneset = GeneSet(name='test')
geneset.save()
geneset.bulk([genes[c] for c in 'someday'])
self.genes = genes
self.geneset = geneset
def _get_or_create_gene(self, record):
target = {}
# Link HGNC id
target['hgnc_id'] = int(record['hgnc_id'])
# Parse and map chromosome
if record['chromosome'] == 'mitochondria':
target['chr_id'] = self.chromosomes['MT']
elif ' and ' in record['chromosome']:
target['chr_id'] = self.chromosomes['XY']
else:
match = self.chrom_re.match(record['chromosome'])
if not match:
log.warning('unable to match gene chromosome from HGNC',
extra={
'hgnc_id': record['hgnc_id'],
'raw_chr': record['chromosome'],
})
return None, False
target['chr_id'] = self.chromosomes[match.groups()[0]]
target['symbol'] = record['approved_symbol'].encode('utf8')
target['name'] = record['approved_name'].encode('utf8')
# If the gene already exists by hgnc_id, fetch it. Next check
# by symbol (in case of a new approved gene) and set the
# hgnc_id. Fallback to creating a new gene
try:
return Gene.objects.get(hgnc_id=target['hgnc_id']), False
except Gene.DoesNotExist:
try:
gene = Gene.objects.get(symbol=target['symbol'])
gene.hgnc_id = target['hgnc_id']
except Gene.DoesNotExist:
gene = Gene(**target)
gene.save()
return gene, True
def get_gene(self, gene_name):
"Get a gene from the cache or attempt to disambiguate or add a new record."
if not gene_name:
return
gene_pk = self.gene_cache.get(gene_name, None)
if gene_pk:
return Gene(pk=gene_pk)
# Attempt to disambiguate, only if this is the only synonym may it be
# associated
potential_genes = list(Gene.objects.filter(synonyms__label__iexact=gene_name).distinct())
if len(potential_genes) == 1:
self.gene_cache[gene_name] = potential_genes[0].pk
return potential_genes[0]
# Only if there are no matches should we create a new record,
# otherwise the synonym war will continue
if len(potential_genes) == 0:
gene = Gene(chr=self.get_chromosome(self.chrom), symbol=gene_name)
gene.save()
self.gene_cache[gene_name] = gene.pk
return gene
def _get_or_create_gene(self, record):
target = {}
# Link HGNC id
target["hgnc_id"] = int(record["hgnc_id"])
# Parse and map chromosome
if record["chromosome"] == "mitochondria":
target["chr_id"] = self.chromosomes["MT"]
elif " and " in record["chromosome"]:
target["chr_id"] = self.chromosomes["XY"]
else:
match = self.chrom_re.match(record["chromosome"])
if not match:
log.warning(
"unable to match gene chromosome from HGNC",
extra={"hgnc_id": record["hgnc_id"], "raw_chr": record["chromosome"]},
)
return None, False
target["chr_id"] = self.chromosomes[match.groups()[0]]
target["symbol"] = record["approved_symbol"].encode("utf8")
target["name"] = record["approved_name"].encode("utf8")
# If the gene already exists by hgnc_id, fetch it. Next check
# by symbol (in case of a new approved gene) and set the
# hgnc_id. Fallback to creating a new gene
try:
return Gene.objects.get(hgnc_id=target["hgnc_id"]), False
except Gene.DoesNotExist:
try:
gene = Gene.objects.get(symbol=target["symbol"])
gene.hgnc_id = target["hgnc_id"]
except Gene.DoesNotExist:
gene = Gene(**target)
gene.save()
return gene, True
def _get_or_create_gene(self, record):
target = {}
# Link HGNC id
target['hgnc_id'] = int(record['hgnc_id'])
# Parse and map chromosome
if record['chromosome'] == 'mitochondria':
target['chr_id'] = self.chromosomes['MT']
elif ' and ' in record['chromosome']:
target['chr_id'] = self.chromosomes['XY']
else:
match = self.chrom_re.match(record['chromosome'])
if not match:
log.warning('unable to match gene chromosome from HGNC', extra={
'hgnc_id': record['hgnc_id'],
'raw_chr': record['chromosome'],
})
return None, False
target['chr_id'] = self.chromosomes[match.groups()[0]]
target['symbol'] = record['approved_symbol'].encode('utf8')
target['name'] = record['approved_name'].encode('utf8')
# If the gene already exists by hgnc_id, fetch it. Next check
# by symbol (in case of a new approved gene) and set the
# hgnc_id. Fallback to creating a new gene
try:
return Gene.objects.get(hgnc_id=target['hgnc_id']), False
except Gene.DoesNotExist:
try:
gene = Gene.objects.get(symbol=target['symbol'])
gene.hgnc_id = target['hgnc_id']
except Gene.DoesNotExist:
gene = Gene(**target)
gene.save()
return gene, True
