def test_parse_ensembl_transcripts(transcripts_handle):
"""Test to parse all ensembl transcripts"""
transcripts = parse_ensembl_transcripts(transcripts_handle)
for transcript in transcripts:
assert transcript['ensembl_transcript_id']
assert transcript['ensembl_gene_id']
def test_parse_ensembl_transcripts(transcripts_handle):
"""Test to parse all ensembl transcripts"""
transcripts = parse_ensembl_transcripts(transcripts_handle)
for transcript in transcripts:
assert transcript['ensembl_transcript_id']
assert transcript['ensembl_gene_id']
def generate_ensembl_transcripts(ensembl_genes, build=None):
"""Generate a file with reduced ensembl gene information
Args:
genes(dict): A dictionary with ensembl_id as key and hgnc_id as value
build(str): What build to use. Defaults to 37
Yields:
print_line(str): Lines from the reduced file
"""
build = build or "37"
ensembl_transcripts = fetch_ensembl_transcripts(build=build)
ensembl_header = [
"Chromosome/scaffold name",
"Gene stable ID",
"Transcript stable ID",
"Transcript start (bp)",
"Transcript end (bp)",
"RefSeq mRNA ID",
"RefSeq mRNA predicted ID",
"RefSeq ncRNA ID",
]
yield "\t".join(ensembl_header)
for tx_info in parse_ensembl_transcripts(ensembl_transcripts):
ens_gene_id = tx_info["ensembl_gene_id"]
if ens_gene_id in ensembl_genes:
print_line = [
tx_info["chrom"],
tx_info["ensembl_gene_id"],
tx_info["ensembl_transcript_id"],
str(tx_info["transcript_start"]),
str(tx_info["transcript_end"]),
tx_info["refseq_mrna"] or "",
tx_info["refseq_mrna_predicted"] or "",
tx_info["refseq_ncrna"] or "",
]
yield "\t".join(print_line)
def generate_ensembl_transcripts(ensembl_genes, build=None):
"""Generate a file with reduced ensembl gene information
Args:
genes(dict): A dictionary with ensembl_id as key and hgnc_id as value
build(str): What build to use. Defaults to 37
Yields:
print_line(str): Lines from the reduced file
"""
build = build or '37'
ensembl_transcripts = fetch_ensembl_transcripts(build=build)
ensembl_header = [
'Chromosome/scaffold name', 'Gene stable ID', 'Transcript stable ID',
'Transcript start (bp)', 'Transcript end (bp)', 'RefSeq mRNA ID',
'RefSeq mRNA predicted ID', 'RefSeq ncRNA ID'
]
yield '\t'.join(ensembl_header)
for tx_info in parse_ensembl_transcripts(ensembl_transcripts):
ens_gene_id = tx_info['ensembl_gene_id']
if ens_gene_id in ensembl_genes:
print_line = [
tx_info['chrom'],
tx_info['ensembl_gene_id'],
tx_info['ensembl_transcript_id'],
str(tx_info['transcript_start']),
str(tx_info['transcript_end']),
tx_info['refseq_mrna'] or '',
tx_info['refseq_mrna_predicted'] or '',
tx_info['refseq_ncrna'] or '',
]
yield '\t'.join(print_line)
def link_genes(ensembl_lines, hgnc_lines, exac_lines, mim2gene_lines,
genemap_lines, hpo_lines):
"""Gather information from different sources and return a gene dict
Extract information collected from a number of sources and combine them
into a gene dict with HGNC symbols as keys.
hgnc_id works as the primary symbol and it is from this source we gather
as much information as possible (hgnc_complete_set.txt)
Coordinates are gathered from ensemble and the entries are linked from hgnc
to ensembl via ENSGID.
From exac the gene intolerance scores are collected, genes are linked to hgnc
via hgnc symbol. This is a unstable symbol since they often change.
Args:
ensembl_lines(iterable(str))
hgnc_lines(iterable(str))
exac_lines(iterable(str))
Yields:
gene(dict): A dictionary with gene information
"""
genes = {}
log.info("Linking genes and transcripts")
# HGNC genes are the main source, these define the gene dataset to use
# Try to use as much information as possible from hgnc
for hgnc_gene in parse_hgnc_genes(hgnc_lines):
hgnc_id = hgnc_gene['hgnc_id']
hgnc_gene['transcripts'] = []
genes[hgnc_id] = hgnc_gene
symbol_to_id = genes_by_alias(genes)
# Parse and add the ensembl gene info
all_genes = {'ensembl': {}, 'symbol': {}}
for transcript in parse_ensembl_transcripts(ensembl_lines):
ensg_symbol = transcript['hgnc_symbol']
ensgid = transcript['ensembl_gene_id']
for id_type, gene_id in [('symbol', ensg_symbol), ('ensembl', ensgid)]:
if gene_id in all_genes[id_type]:
all_genes[id_type][gene_id].append(transcript)
else:
all_genes[id_type][gene_id] = [transcript]
log.info("Add ensembl info")
# Add gene coordinates and transcript info for hgnc genes:
for gene_info in genes.values():
ensgid = gene_info['ensembl_gene_id']
ensg_symbol = gene_info['hgnc_symbol']
for id_type, gene_id in [('ensembl', ensgid), ('symbol', ensg_symbol)]:
if gene_id:
if gene_id in all_genes[id_type]:
add_ensembl_info(gene_info, all_genes[id_type][gene_id])
ensgid = 'ADDED'
break
log.info("Add exac pli scores")
for exac_gene in parse_exac_genes(exac_lines):
hgnc_symbol = exac_gene['hgnc_symbol'].upper()
pli_score = exac_gene['pli_score']
if hgnc_symbol in symbol_to_id:
hgnc_id_info = symbol_to_id[hgnc_symbol]
# If we have the true id we know ot os correct
if hgnc_id_info['true_id']:
hgnc_id = hgnc_id_info['true_id']
genes[hgnc_id]['pli_score'] = pli_score
# Otherwise we loop over the ids and add pli score if it
# is not already added
else:
for hgnc_id in hgnc_id_info['ids']:
gene_info = genes[hgnc_id]
if not gene_info.get('pli_score'):
gene_info['pli_score'] = pli_score
log.info("Add omim info")
omim_genes = get_mim_genes(genemap_lines, mim2gene_lines)
for hgnc_symbol in omim_genes:
omim_info = omim_genes[hgnc_symbol]
inheritance = omim_info.get('inheritance', set())
if hgnc_symbol in symbol_to_id:
hgnc_id_info = symbol_to_id[hgnc_symbol]
# If we have the true id we know it is correct
if hgnc_id_info['true_id']:
hgnc_id = hgnc_id_info['true_id']
gene_info = genes[hgnc_id]
# Update the omim id to the one found in omim
gene_info['omim_id'] = omim_info['mim_number']
gene_info['inheritance_models'] = list(inheritance)
gene_info['phenotypes'] = omim_info.get('phenotypes', [])
else:
for hgnc_id in hgnc_id_info['ids']:
gene_info = genes[hgnc_id]
if not gene_info.get('omim_id'):
gene_info['omim_id'] = omim_info['mim_number']
if not gene_info.get('inheritance_models'):
gene_info['inheritance_models'] = list(inheritance)
if not gene_info.get('phenotypes'):
gene_info['phenotypes'] = omim_info.get('phenotypes', [])
log.info("Add incomplete penetrance info")
for hgnc_symbol in get_incomplete_penetrance_genes(hpo_lines):
if hgnc_symbol in symbol_to_id:
hgnc_id_info = symbol_to_id[hgnc_symbol]
# If we have the true id we know ot os correct
if hgnc_id_info['true_id']:
hgnc_id = hgnc_id_info['true_id']
genes[hgnc_id]['incomplete_penetrance'] = True
# Otherwise we loop over the ids and add incomplete penetrance if it
# is not already added
else:
for hgnc_id in hgnc_id_info['ids']:
gene_info = genes[hgnc_id]
if not 'incomplete_penetrance' in gene_info:
gene_info['incomplete_penetrance'] = True
return genes
def transcripts(request, transcripts_handle):
"""Get the parsed ensembl transcripts"""
print('')
return parse_ensembl_transcripts(transcripts_handle)