def parse_transcript(gene_obj, tx_obj, build=None):
"""Parse variant gene transcript (VEP)."""
build = build or 37
add_tx_links(tx_obj, build)
if tx_obj.get('refseq_id'):
gene_name = (gene_obj['common']['hgnc_symbol'] if gene_obj['common'] else
gene_obj['hgnc_id'])
tx_obj['change_str'] = transcript_str(tx_obj, gene_name)
def gene(store, hgnc_id):
"""Parse information about a gene."""
res = {
'builds': {
'37': None,
'38': None
},
'symbol': None,
'description': None,
'ensembl_id': None,
'record': None
}
for build in res['builds']:
record = store.hgnc_gene(hgnc_id, build=build)
if record:
record[
'position'] = "{this[chromosome]}:{this[start]}-{this[end]}".format(
this=record)
res['aliases'] = record['aliases']
res['hgnc_id'] = record['hgnc_id']
res['description'] = record['description']
res['builds'][build] = record
res['symbol'] = record['hgnc_symbol']
res['description'] = record['description']
res['entrez_id'] = record.get('entrez_id')
res['pli_score'] = record.get('pli_score')
add_gene_links(record, int(build))
res['omim_id'] = record.get('omim_id')
res['incomplete_penetrance'] = record.get('incomplete_penetrance',
False)
res['inheritance_models'] = record.get('inheritance_models', [])
for transcript in record['transcripts']:
transcript['position'] = (
"{this[chrom]}:{this[start]}-{this[end]}".format(
this=transcript))
add_tx_links(transcript, build)
for phenotype in record.get('phenotypes', []):
phenotype['omim_link'] = omim(phenotype.get('mim_number'))
if not res['record']:
res['record'] = record
pp(record)
# If none of the genes where found
if not any(res.values()):
raise ValueError
return res
def gene(store, hgnc_id):
"""Parse information about a gene."""
res = {
"builds": {
"37": None,
"38": None
},
"symbol": None,
"description": None,
"ensembl_id": None,
"record": None,
}
for build in res["builds"]:
record = store.hgnc_gene(hgnc_id, build=build)
if record:
record[
"position"] = "{this[chromosome]}:{this[start]}-{this[end]}".format(
this=record)
res["aliases"] = record["aliases"]
res["hgnc_id"] = record["hgnc_id"]
res["description"] = record["description"]
res["builds"][build] = record
res["symbol"] = record["hgnc_symbol"]
res["description"] = record["description"]
res["entrez_id"] = record.get("entrez_id")
res["pli_score"] = record.get("pli_score")
add_gene_links(record, int(build))
res["omim_id"] = record.get("omim_id")
res["incomplete_penetrance"] = record.get("incomplete_penetrance",
False)
res["inheritance_models"] = record.get("inheritance_models", [])
for transcript in record["transcripts"]:
transcript[
"position"] = "{this[chrom]}:{this[start]}-{this[end]}".format(
this=transcript)
add_tx_links(transcript, build)
for phenotype in record.get("phenotypes", []):
phenotype["omim_link"] = omim(phenotype.get("mim_number"))
if not res["record"]:
res["record"] = record
# If none of the genes where found
if not any(res.values()):
raise ValueError
return res
def gene(store, hgnc_id):
"""Parse information about a gene."""
res = {'builds': {'37': None, '38': None}, 'symbol': None, 'description': None, 'ensembl_id': None, 'record': None}
for build in res['builds']:
record = store.hgnc_gene(hgnc_id, build=build)
if record:
record['position'] = "{this[chromosome]}:{this[start]}-{this[end]}".format(this=record)
res['aliases'] = record['aliases']
res['hgnc_id'] = record['hgnc_id']
res['description'] = record['description']
res['builds'][build] = record
res['symbol'] = record['hgnc_symbol']
res['description'] = record['description']
res['entrez_id'] = record.get('entrez_id')
res['pli_score'] = record.get('pli_score')
add_gene_links(record, int(build))
res['omim_id'] = record.get('omim_id')
res['incomplete_penetrance'] = record.get('incomplete_penetrance',False)
res['inheritance_models'] = record.get('inheritance_models',[])
for transcript in record['transcripts']:
transcript['position'] = ("{this[chrom]}:{this[start]}-{this[end]}"
.format(this=transcript))
add_tx_links(transcript, build)
for phenotype in record.get('phenotypes',[]):
phenotype['omim_link'] = omim(phenotype.get('mim_number'))
if not res['record']:
res['record'] = record
# If none of the genes where found
if not any(res.values()):
raise ValueError
return res
def update_transcripts_information(variant_gene,
hgnc_gene,
variant_obj,
genome_build=None):
"""Collect tx info from the hgnc gene and panels and update variant transcripts
Since the hgnc information are continuously being updated we need to run this each time a
variant is fetched.
This function will:
- Add a dictionary with tx_id -> tx_info to the hgnc variant
- Add information from the panel
- Adds a list of refseq transcripts
Args:
variant_gene(dict): the gene information from the variant
hgnc_gene(dict): the hgnc gene information
varaiant_obj(scout.models.Variant)
"""
genome_build = genome_build or "37"
disease_associated_no_version = variant_gene.get(
"disease_associated_no_version", set())
# Create a dictionary with transcripts information
# Use ensembl transcript id as keys
transcripts_dict = {}
# Add transcript information from the hgnc gene
for transcript in hgnc_gene.get("transcripts", []):
tx_id = transcript["ensembl_transcript_id"]
transcripts_dict[tx_id] = transcript
# Add the transcripts to the gene object
hgnc_gene["transcripts_dict"] = transcripts_dict
hgnc_symbol = hgnc_gene["hgnc_symbol"]
refseq_transcripts = []
# First loop over the variants transcripts
for transcript in variant_gene.get("transcripts", []):
tx_id = transcript["transcript_id"]
hgnc_transcript = transcripts_dict.get(tx_id)
# If the tx does not exist in ensembl anymore we skip it
if not hgnc_transcript:
continue
# Check in the common information if it is a primary transcript
if hgnc_transcript.get("is_primary"):
transcript["is_primary"] = True
# Add the transcript links
add_tx_links(transcript, genome_build)
# If the transcript has a ref seq identifier we add that
# to the variants transcript
refseq_id = hgnc_transcript.get("refseq_id")
if not refseq_id:
continue
transcript["refseq_id"] = refseq_id
variant_obj["has_refseq"] = True
refseq_transcripts.append(transcript)
# Check if the refseq id are disease associated
if refseq_id in disease_associated_no_version:
transcript["is_disease_associated"] = True
# Since a ensemble transcript can have multiple refseq identifiers we add all of
# those
transcript["refseq_identifiers"] = hgnc_transcript.get(
"refseq_identifiers", [])
transcript["change_str"] = transcript_str(transcript, hgnc_symbol)
variant_gene["primary_transcripts"] = refseq_transcripts