def from_assembly(cls,
assembly: Assembly,
alt_aln_method: str = 'splign') -> 'HgvsMachinery':
"""
Initializes `biocommons/hgvs` machinery with Universal Transcript Archive (UTA) data provider.
:param alt_aln_method: alignment method (default 'splign').
:param assembly: determines the assembly build.
"""
data_provider = uta.connect()
assembly_name = ASSEMBLY__HGVS_ASSEMBLY_NAME[assembly]
accession__contig = data_provider.get_assembly_map(
assembly_name=assembly_name)
return cls(assembly_mapper=AssemblyMapper(
hdp=data_provider,
assembly_name=assembly_name,
alt_aln_method=alt_aln_method,
add_gene_symbol=False,
prevalidation_level='NONE',
normalize=True,
replace_reference=False),
normalizer_3p=Normalizer(hdp=data_provider,
shuffle_direction=3),
normalizer_5p=Normalizer(hdp=data_provider,
shuffle_direction=5),
accession__contig=accession__contig,
contig__accession={
contig: acc
for acc, contig in accession__contig.items()
})
def annotator(annotation):
# 连接UTA, 创建Mapper、Parser、Normalizer
hdp = connect()
am = AssemblyMapper(hdp,
assembly_name=annotation,
alt_aln_method='splign',
replace_reference=True)
hp = Parser()
hn = Normalizer(hdp)
return am, hp, hn, hdp
HGVS_SEQREPO_DIR environment variables but is otherwise not
configurable by the caller.
"""
from hgvs import __version__, global_config # noqa: F401
from hgvs.assemblymapper import AssemblyMapper
from hgvs.dataproviders.uta import connect
from hgvs.normalizer import Normalizer
from hgvs.parser import Parser
from hgvs.validator import Validator
from hgvs.variantmapper import VariantMapper
# provide standard abbreviated, short, and long names for instances
hp = parser = hgvs_parser = Parser()
hdp = hgvs_data_provider = connect()
vm = variant_mapper = hgvs_variant_mapper = VariantMapper(hgvs_data_provider)
am37 = hgvs_assembly_mapper_37 = AssemblyMapper(hgvs_data_provider,
assembly_name='GRCh37')
am38 = projector = hgvs_assembly_mapper_38 = AssemblyMapper(
hgvs_data_provider, assembly_name='GRCh38')
hn = normalizer = hgvs_normalizer = Normalizer(hgvs_data_provider)
hv = validator = hgvs_validator = Validator(hgvs_data_provider)
# functionalized forms of common methods
parse = parser.parse
normalize = normalizer.normalize
validate = validator.validate
c_to_g = projector.c_to_g
c_to_n = projector.c_to_n
c_to_p = projector.c_to_p
def main(args):
options = parse_args()
brcaFile = options.inBRCA
hg18_fa = options.inHg18
hg19_fa = options.inHg19
hg38_fa = options.inHg38
refSeq18 = options.inRefSeq18
refSeq19 = options.inRefSeq19
refSeq38 = options.inRefSeq38
outputFile = options.outBRCA
calcProtein = options.calcProtein
artifacts_dir = options.artifacts_dir
if not os.path.exists(artifacts_dir):
os.makedirs(artifacts_dir)
log_file_path = artifacts_dir + "brca-pseudonym-generator.log"
logging.basicConfig(filename=log_file_path,
filemode="w",
level=logging.DEBUG)
hdp = hgvs_dataproviders_uta.connect()
variantmapper = hgvs_variantmapper.EasyVariantMapper(hdp)
hgvsparser = hgvs_parser.Parser()
genome36 = SequenceFileDB(hg18_fa.name)
genome37 = SequenceFileDB(hg19_fa.name)
genome38 = SequenceFileDB(hg38_fa.name)
transcripts36 = pyhgvs_utils.read_transcripts(refSeq18)
transcripts37 = pyhgvs_utils.read_transcripts(refSeq19)
transcripts38 = pyhgvs_utils.read_transcripts(refSeq38)
def get_transcript36(name):
return transcripts36.get(name)
def get_transcript37(name):
return transcripts37.get(name)
def get_transcript38(name):
return transcripts38.get(name)
hgvsG36ColumnName = 'Genomic_Coordinate_hg36'
hgvsG37ColumnName = 'Genomic_Coordinate_hg37'
hgvsG38ColumnName = 'Genomic_Coordinate_hg38'
refSeqColumnName = 'Reference_Sequence'
hgvsCDNAColumnName = 'HGVS_cDNA'
hgvsCDNALOVDColumnName = 'HGVS_cDNA_LOVD'
hgvsPColumnName = 'HGVS_Protein'
# Set up header for output file
input_file = csv.reader(brcaFile, delimiter='\t')
output_file = csv.writer(outputFile, delimiter='\t')
input_header_row = input_file.next()
# The following new columns will contain data generated by this file
new_columns_to_append = [
"pyhgvs_Genomic_Coordinate_36", "pyhgvs_Genomic_Coordinate_37",
"pyhgvs_Genomic_Coordinate_38", "pyhgvs_Hg37_Start", "pyhgvs_Hg37_End",
"pyhgvs_Hg36_Start", "pyhgvs_Hg36_End", "pyhgvs_cDNA", "pyhgvs_Protein"
]
output_header_row = input_header_row + new_columns_to_append
output_file.writerow(output_header_row)
# Store indexes of the relevant columns
hgvsG36Index = input_header_row.index(hgvsG36ColumnName)
hgvsG37Index = input_header_row.index(hgvsG37ColumnName)
hgvsG38Index = input_header_row.index(hgvsG38ColumnName)
refSeqIndex = input_header_row.index(refSeqColumnName)
hgvsCDNAIndex = input_header_row.index(hgvsCDNAColumnName)
hgvsPIndex = input_header_row.index(hgvsPColumnName)
hgvsCDNALOVDIndex = input_header_row.index(hgvsCDNALOVDColumnName)
geneSymbolIndex = input_header_row.index("Gene_Symbol")
synonymIndex = input_header_row.index("Synonyms")
refSeqBRCA1Transcripts = [
'NM_007294.2', 'NM_007300.3', 'NM_007299.3', 'NM_007298.3',
'NM_007297.3', 'U14680.1'
]
refSeqBRCA2Transcripts = ['U43746.1']
for line in input_file:
if line[geneSymbolIndex] == 'BRCA1':
line[refSeqIndex] = 'NM_007294.3'
elif line[geneSymbolIndex] == 'BRCA2':
line[refSeqIndex] = 'NM_000059.3'
# Store for reference and debugging
oldHgvsGenomic38 = line[refSeqIndex] + ':' + line[hgvsG38Index].split(
',')[0]
chrom38 = line[input_header_row.index("Chr")]
offset38 = line[input_header_row.index("Pos")]
ref38 = line[input_header_row.index("Ref")]
alt38 = line[input_header_row.index("Alt")]
# Edge cases to correct variant string formats for indels in order to be accepted by the counsyl parser
if ref38 == '-': ref38 = ''
if alt38 == '-': alt38 = ''
if alt38 == 'None': alt38 = ''
transcript38 = get_transcript38(line[refSeqIndex])
transcript37 = get_transcript37(line[refSeqIndex])
transcript36 = get_transcript36(line[refSeqIndex])
# Normalize hgvs cdna string to fit what the counsyl hgvs parser determines to be the correct format
if transcript38 is None:
print("ERROR: could not parse transcript38 for variant: %s \n" %
(line))
continue
cdna_coord = str(
pyhgvs.format_hgvs_name("chr" + chrom38,
int(offset38),
ref38,
alt38,
genome38,
transcript38,
use_gene=False,
max_allele_length=100))
chrom38, offset38, ref38, alt38 = pyhgvs.parse_hgvs_name(
cdna_coord, genome38, get_transcript=get_transcript38)
chrom37, offset37, ref37, alt37 = pyhgvs.parse_hgvs_name(
cdna_coord, genome37, get_transcript=get_transcript37)
chrom36, offset36, ref36, alt36 = pyhgvs.parse_hgvs_name(
cdna_coord, genome36, get_transcript=get_transcript36)
# Generate transcript hgvs cdna synonym string
if line[synonymIndex] == "-":
synonymString = []
elif line[synonymIndex] == "":
synonymString = []
else:
synonymString = line[synonymIndex].split(",")
if line[geneSymbolIndex] == 'BRCA1':
for transcriptName in refSeqBRCA1Transcripts:
transcript38 = get_transcript38(transcriptName)
cdna_synonym = str(
pyhgvs.format_hgvs_name(chrom38,
int(offset38),
ref38,
alt38,
genome38,
transcript38,
use_gene=False,
max_allele_length=100))
synonymString.append(cdna_synonym)
elif line[geneSymbolIndex] == 'BRCA2':
for transcriptName in refSeqBRCA2Transcripts:
transcript38 = get_transcript38(transcriptName)
cdna_synonym = str(
pyhgvs.format_hgvs_name(chrom38,
int(offset38),
ref38,
alt38,
genome38,
transcript38,
use_gene=False,
max_allele_length=100))
synonymString.append(cdna_synonym)
# Add hgvs_cDNA values from LOVD to synonyms if not already present
for cdna_coord_LOVD in line[hgvsCDNALOVDIndex].split(','):
# Skip if blank
if cdna_coord_LOVD == "-" or cdna_coord_LOVD is None or cdna_coord_LOVD == "":
continue
# Don't add to synonyms if main hgvs_cDNA field is already equivalent to hgvs_cDNA value from LOVD
cdna_coord_LOVD_for_comparison = cdna_coord_LOVD.split(':')[1]
if cdna_coord_LOVD_for_comparison in line[hgvsCDNAIndex]:
continue
chrom38LOVD, offset38LOVD, ref38LOVD, alt38LOVD = pyhgvs.parse_hgvs_name(
cdna_coord_LOVD, genome38, get_transcript=get_transcript38)
if line[geneSymbolIndex] == 'BRCA1':
for transcriptName in refSeqBRCA1Transcripts:
transcript38 = get_transcript38(transcriptName)
cdna_synonym = str(
pyhgvs.format_hgvs_name(chrom38LOVD,
int(offset38LOVD),
ref38LOVD,
alt38LOVD,
genome38,
transcript38,
use_gene=False,
max_allele_length=100))
if cdna_synonym not in synonymString:
synonymString.append(cdna_synonym)
elif line[geneSymbolIndex] == 'BRCA2':
for transcriptName in refSeqBRCA2Transcripts:
transcript38 = get_transcript38(transcriptName)
cdna_synonym = str(
pyhgvs.format_hgvs_name(chrom38LOVD,
int(offset38LOVD),
ref38LOVD,
alt38LOVD,
genome38,
transcript38,
use_gene=False,
max_allele_length=100))
if cdna_synonym not in synonymString:
synonymString.append(cdna_synonym)
if calcProtein == True:
try:
var_c1 = hgvsparser.parse_hgvs_variant(cdna_coord)
protein_coord = variantmapper.c_to_p(var_c1)
except hgvs.exceptions.HGVSParseError as e:
template = "An exception of type {0} occured. Arguments:\n{1!r}"
message = template.format(type(ex).__name__, ex.args)
genomicChange = '{0}:g.{1}:{2}>{3}'.format(
chrom38, offset38, ref38, alt38)
print('hgvs.exceptions.HGVSParseError: ', e)
print('Original GRCh38 Genomic Coordinate: ', oldHgvsGenomic38)
print('GRCh38 Genomic change: ', genomicChange)
logging.error(message)
logging.error(line)
logging.error('Proposed GRCh38 Genomic change for error: %s',
genomicChange)
# Catch parse errors thrown by ometa.runtime.ParseError.
except ParseError as ex:
template = "An exception of type {0} occured. Arguments:\n{1!r}"
message = template.format(type(ex).__name__, ex.args)
genomicChange = '{0}:g.{1}:{2}>{3}'.format(
chrom38, offset38, ref38, alt38)
print(message)
print('ometa.runtime.ParseError', ex)
print('Original GRCh38 Genomic Coordinate: ', oldHgvsGenomic38)
print('GRCh38 Genomic change: ', genomicChange)
logging.error(message)
logging.error(line)
logging.error('Proposed GRCh38 Genomic change for error: %s',
genomicChange)
# Add empty data for each new column to prepare for data insertion by index
for i in range(len(new_columns_to_append)):
line.append('-')
line[output_header_row.index(
"pyhgvs_Genomic_Coordinate_36")] = '{0}:g.{1}:{2}>{3}'.format(
chrom36, offset36, ref36, alt36)
line[output_header_row.index(
"pyhgvs_Genomic_Coordinate_37")] = '{0}:g.{1}:{2}>{3}'.format(
chrom37, offset37, ref37, alt37)
line[output_header_row.index(
"pyhgvs_Genomic_Coordinate_38")] = '{0}:g.{1}:{2}>{3}'.format(
chrom38, offset38, ref38, alt38)
line[output_header_row.index("pyhgvs_Hg37_Start")] = str(offset37)
line[output_header_row.index("pyhgvs_Hg37_End")] = str(
int(offset37) + len(ref38) - 1)
line[output_header_row.index("pyhgvs_Hg36_Start")] = str(offset36)
line[output_header_row.index("pyhgvs_Hg36_End")] = str(
int(offset36) + len(ref38) - 1)
line[output_header_row.index("pyhgvs_cDNA")] = '{0}'.format(cdna_coord)
if calcProtein == True:
line[output_header_row.index("pyhgvs_Protein")] = '{0}'.format(
str(protein_coord))
line[synonymIndex] = ','.join(synonymString)
output_file.writerow(line)
hg18_fa.close()
hg19_fa.close()
hg38_fa.close()
refSeq18.close()
refSeq19.close()
refSeq38.close()
def main(args):
options = parse_args()
brcaFile = options.inBRCA
hg18_fa = options.inHg18
hg19_fa = options.inHg19
hg38_fa = options.inHg38
refSeq18 = options.inRefSeq18
refSeq19 = options.inRefSeq19
refSeq38 = options.inRefSeq38
outputFile = options.outBRCA
calcProtein = options.calcProtein
artifacts_dir = options.artifacts_dir
if not os.path.exists(artifacts_dir):
os.makedirs(artifacts_dir)
log_file_path = artifacts_dir + "brca-pseudonym-generator.log"
logging.basicConfig(filename=log_file_path, filemode="w", level=logging.DEBUG)
hdp = hgvs_dataproviders_uta.connect()
variantmapper = hgvs_variantmapper.EasyVariantMapper(hdp)
hgvsparser = hgvs_parser.Parser()
genome36 = SequenceFileDB(hg18_fa.name)
genome37 = SequenceFileDB(hg19_fa.name)
genome38 = SequenceFileDB(hg38_fa.name)
transcripts36 = pyhgvs_utils.read_transcripts(refSeq18)
transcripts37 = pyhgvs_utils.read_transcripts(refSeq19)
transcripts38 = pyhgvs_utils.read_transcripts(refSeq38)
def get_transcript36(name):
return transcripts36.get(name)
def get_transcript37(name):
return transcripts37.get(name)
def get_transcript38(name):
return transcripts38.get(name)
hgvsG36ColumnName = 'Genomic_Coordinate_hg36'
hgvsG37ColumnName = 'Genomic_Coordinate_hg37'
hgvsG38ColumnName = 'Genomic_Coordinate_hg38'
refSeqColumnName = 'Reference_Sequence'
hgvsCDNAColumnName = 'HGVS_cDNA'
hgvsCDNALOVDColumnName = 'HGVS_cDNA_LOVD'
hgvsPColumnName = 'HGVS_Protein'
# Set up header for output file
input_file = csv.reader(brcaFile, delimiter='\t')
output_file = csv.writer(outputFile, delimiter='\t')
input_header_row = input_file.next()
# The following new columns will contain data generated by this file
new_columns_to_append = ["pyhgvs_Genomic_Coordinate_36", "pyhgvs_Genomic_Coordinate_37",
"pyhgvs_Genomic_Coordinate_38", "pyhgvs_Hg37_Start", "pyhgvs_Hg37_End",
"pyhgvs_Hg36_Start", "pyhgvs_Hg36_End", "pyhgvs_cDNA", "pyhgvs_Protein"]
output_header_row = input_header_row + new_columns_to_append
output_file.writerow(output_header_row)
# Store indexes of the relevant columns
hgvsG36Index = input_header_row.index(hgvsG36ColumnName)
hgvsG37Index = input_header_row.index(hgvsG37ColumnName)
hgvsG38Index = input_header_row.index(hgvsG38ColumnName)
refSeqIndex = input_header_row.index(refSeqColumnName)
hgvsCDNAIndex = input_header_row.index(hgvsCDNAColumnName)
hgvsPIndex = input_header_row.index(hgvsPColumnName)
hgvsCDNALOVDIndex = input_header_row.index(hgvsCDNALOVDColumnName)
geneSymbolIndex = input_header_row.index("Gene_Symbol")
synonymIndex = input_header_row.index("Synonyms")
refSeqBRCA1Transcripts = ['NM_007294.2', 'NM_007300.3', 'NM_007299.3', 'NM_007298.3', 'NM_007297.3', 'U14680.1']
refSeqBRCA2Transcripts = ['U43746.1']
for line in input_file:
if line[geneSymbolIndex] == 'BRCA1':
line[refSeqIndex] = 'NM_007294.3'
elif line[geneSymbolIndex] == 'BRCA2':
line[refSeqIndex] = 'NM_000059.3'
# Store for reference and debugging
oldHgvsGenomic38 = line[refSeqIndex] + ':' + line[hgvsG38Index].split(',')[0]
chrom38 = line[input_header_row.index("Chr")]
offset38 = line[input_header_row.index("Pos")]
ref38 = line[input_header_row.index("Ref")]
alt38 = line[input_header_row.index("Alt")]
# Edge cases to correct variant string formats for indels in order to be accepted by the counsyl parser
if ref38 == '-': ref38 = ''
if alt38 == '-': alt38 = ''
if alt38 == 'None': alt38 = ''
transcript38 = get_transcript38(line[refSeqIndex])
transcript37 = get_transcript37(line[refSeqIndex])
transcript36 = get_transcript36(line[refSeqIndex])
# Normalize hgvs cdna string to fit what the counsyl hgvs parser determines to be the correct format
if transcript38 is None:
print("ERROR: could not parse transcript38 for variant: %s \n" % (line))
continue
cdna_coord = str(pyhgvs.format_hgvs_name("chr" + chrom38, int(offset38), ref38, alt38, genome38, transcript38, use_gene=False, max_allele_length=100))
chrom38, offset38, ref38, alt38 = pyhgvs.parse_hgvs_name(cdna_coord, genome38, get_transcript=get_transcript38)
chrom37, offset37, ref37, alt37 = pyhgvs.parse_hgvs_name(cdna_coord, genome37, get_transcript=get_transcript37)
chrom36, offset36, ref36, alt36 = pyhgvs.parse_hgvs_name(cdna_coord, genome36, get_transcript=get_transcript36)
# Generate transcript hgvs cdna synonym string
if line[synonymIndex] == "-":
synonymString = []
elif line[synonymIndex] == "":
synonymString = []
else:
synonymString = line[synonymIndex].split(",")
if line[geneSymbolIndex] == 'BRCA1':
for transcriptName in refSeqBRCA1Transcripts:
transcript38 = get_transcript38(transcriptName)
cdna_synonym = str(pyhgvs.format_hgvs_name(chrom38, int(offset38), ref38, alt38, genome38, transcript38, use_gene=False, max_allele_length=100))
synonymString.append(cdna_synonym)
elif line[geneSymbolIndex] == 'BRCA2':
for transcriptName in refSeqBRCA2Transcripts:
transcript38 = get_transcript38(transcriptName)
cdna_synonym = str(pyhgvs.format_hgvs_name(chrom38, int(offset38), ref38, alt38, genome38, transcript38, use_gene=False, max_allele_length=100))
synonymString.append(cdna_synonym)
# Add hgvs_cDNA values from LOVD to synonyms if not already present
for cdna_coord_LOVD in line[hgvsCDNALOVDIndex].split(','):
# Skip if blank
if cdna_coord_LOVD == "-" or cdna_coord_LOVD is None or cdna_coord_LOVD == "":
continue
# Don't add to synonyms if main hgvs_cDNA field is already equivalent to hgvs_cDNA value from LOVD
cdna_coord_LOVD_for_comparison = cdna_coord_LOVD.split(':')[1]
if cdna_coord_LOVD_for_comparison in line[hgvsCDNAIndex]:
continue
chrom38LOVD, offset38LOVD, ref38LOVD, alt38LOVD = pyhgvs.parse_hgvs_name(cdna_coord_LOVD, genome38, get_transcript=get_transcript38)
if line[geneSymbolIndex] == 'BRCA1':
for transcriptName in refSeqBRCA1Transcripts:
transcript38 = get_transcript38(transcriptName)
cdna_synonym = str(pyhgvs.format_hgvs_name(chrom38LOVD, int(offset38LOVD), ref38LOVD, alt38LOVD, genome38, transcript38, use_gene=False, max_allele_length=100))
if cdna_synonym not in synonymString:
synonymString.append(cdna_synonym)
elif line[geneSymbolIndex] == 'BRCA2':
for transcriptName in refSeqBRCA2Transcripts:
transcript38 = get_transcript38(transcriptName)
cdna_synonym = str(pyhgvs.format_hgvs_name(chrom38LOVD, int(offset38LOVD), ref38LOVD, alt38LOVD, genome38, transcript38, use_gene=False, max_allele_length=100))
if cdna_synonym not in synonymString:
synonymString.append(cdna_synonym)
if calcProtein == True:
try:
var_c1 = hgvsparser.parse_hgvs_variant(cdna_coord)
protein_coord = variantmapper.c_to_p(var_c1)
except hgvs.exceptions.HGVSParseError as e:
template = "An exception of type {0} occured. Arguments:\n{1!r}"
message = template.format(type(ex).__name__, ex.args)
genomicChange = '{0}:g.{1}:{2}>{3}'.format(chrom38, offset38, ref38, alt38)
print('hgvs.exceptions.HGVSParseError: ', e)
print('Original GRCh38 Genomic Coordinate: ', oldHgvsGenomic38)
print('GRCh38 Genomic change: ', genomicChange)
logging.error(message)
logging.error(line)
logging.error('Proposed GRCh38 Genomic change for error: %s', genomicChange)
# Catch parse errors thrown by ometa.runtime.ParseError.
except ParseError as ex:
template = "An exception of type {0} occured. Arguments:\n{1!r}"
message = template.format(type(ex).__name__, ex.args)
genomicChange = '{0}:g.{1}:{2}>{3}'.format(chrom38, offset38, ref38, alt38)
print(message)
print('ometa.runtime.ParseError', ex)
print('Original GRCh38 Genomic Coordinate: ', oldHgvsGenomic38)
print('GRCh38 Genomic change: ', genomicChange)
logging.error(message)
logging.error(line)
logging.error('Proposed GRCh38 Genomic change for error: %s', genomicChange)
# Add empty data for each new column to prepare for data insertion by index
for i in range(len(new_columns_to_append)):
line.append('-')
line[output_header_row.index("pyhgvs_Genomic_Coordinate_36")] = '{0}:g.{1}:{2}>{3}'.format(chrom36,offset36,ref36,alt36)
line[output_header_row.index("pyhgvs_Genomic_Coordinate_37")] = '{0}:g.{1}:{2}>{3}'.format(chrom37,offset37,ref37,alt37)
line[output_header_row.index("pyhgvs_Genomic_Coordinate_38")] = '{0}:g.{1}:{2}>{3}'.format(chrom38,offset38,ref38,alt38)
line[output_header_row.index("pyhgvs_Hg37_Start")] = str(offset37)
line[output_header_row.index("pyhgvs_Hg37_End")] = str(int(offset37) + len(ref38) - 1)
line[output_header_row.index("pyhgvs_Hg36_Start")] = str(offset36)
line[output_header_row.index("pyhgvs_Hg36_End")] = str(int(offset36) + len(ref38) - 1)
line[output_header_row.index("pyhgvs_cDNA")] = '{0}'.format(cdna_coord)
if calcProtein == True:
line[output_header_row.index("pyhgvs_Protein")] = '{0}'.format(str(protein_coord))
line[synonymIndex] = ','.join(synonymString)
output_file.writerow(line)
hg18_fa.close()
hg19_fa.close()
hg38_fa.close()
refSeq18.close()
refSeq19.close()
refSeq38.close()
def main(args):
options = parse_args()
brcaFile = options.inBRCA
hg18_fa = options.inHg18
hg19_fa = options.inHg19
hg38_fa = options.inHg38
refSeq18 = options.inRefSeq18
refSeq19 = options.inRefSeq19
refSeq38 = options.inRefSeq38
outputFile = options.outBRCA
calcProtein = options.calcProtein
hdp = hgvs_dataproviders_uta.connect()
variantmapper = hgvs_variantmapper.EasyVariantMapper(hdp)
hgvsparser = hgvs_parser.Parser()
genome36 = SequenceFileDB(hg18_fa.name)
genome37 = SequenceFileDB(hg19_fa.name)
genome38 = SequenceFileDB(hg38_fa.name)
transcripts36 = pyhgvs_utils.read_transcripts(refSeq18)
transcripts37 = pyhgvs_utils.read_transcripts(refSeq19)
transcripts38 = pyhgvs_utils.read_transcripts(refSeq38)
def get_transcript36(name):
return transcripts36.get(name)
def get_transcript37(name):
return transcripts37.get(name)
def get_transcript38(name):
return transcripts38.get(name)
hgvsG36ColumnName = 'Genomic_Coordinate_hg36'
hgvsG37ColumnName = 'Genomic_Coordinate_hg37'
hgvsG38ColumnName = 'Genomic_Coordinate_hg38'
refSeqColumnName = 'Reference_Sequence'
hgvsCDNAColumnName = 'HGVS_cDNA'
hgvsPColumnName = 'HGVS_Protein'
labelLine = brcaFile.readline().rstrip().split('\t')
writeLine = '\t'.join(labelLine) + '\n'
outputFile.writelines(writeLine)
# Store indexes of the relevant columns
hgvsG36Index = labelLine.index(hgvsG36ColumnName)
hgvsG37Index = labelLine.index(hgvsG37ColumnName)
hgvsG38Index = labelLine.index(hgvsG38ColumnName)
refSeqIndex = labelLine.index(refSeqColumnName)
hgvsCDNAIndex = labelLine.index(hgvsCDNAColumnName)
hgvsPIndex = labelLine.index(hgvsPColumnName)
geneSymbolIndex = labelLine.index("Gene_Symbol")
synonymIndex = labelLine.index("Synonyms")
refSeqBRCA1Transcripts = [
'NM_007294.2', 'NM_007300.3', 'NM_007299.3', 'NM_007298.3',
'NM_007297.3', 'U14680.1'
]
refSeqBRCA2Transcripts = ['U43746.1']
for line in brcaFile:
parsedLine = line.rstrip().split('\t')
if parsedLine[geneSymbolIndex] == 'BRCA1':
parsedLine[refSeqIndex] = 'NM_007294.3'
elif parsedLine[geneSymbolIndex] == 'BRCA2':
parsedLine[refSeqIndex] = 'NM_000059.3'
# Format genomic variant position strings to contain relevant refseq strings
oldHgvsGenomic36 = parsedLine[refSeqIndex] + ':' + parsedLine[
hgvsG36Index]
oldHgvsGenomic37 = parsedLine[refSeqIndex] + ':' + parsedLine[
hgvsG37Index]
oldHgvsGenomic38 = parsedLine[refSeqIndex] + ':' + parsedLine[
hgvsG38Index].split(',')[0]
oldHgvsCDNA = parsedLine[refSeqIndex] + ':' + parsedLine[hgvsCDNAIndex]
chrom38 = oldHgvsGenomic38.split(':')[1]
offset38 = oldHgvsGenomic38.split(':')[2]
ref38 = oldHgvsGenomic38.split(':')[3].split('>')[0]
alt38 = oldHgvsGenomic38.split(':')[3].split('>')[1]
# Edge cases to correct variant string formats for indels in order to be accepted by the counsyl parser
if ref38 == '-': ref38 = ''
if alt38 == '-': alt38 = ''
if alt38 == 'None': alt38 = ''
transcript38 = get_transcript38(parsedLine[refSeqIndex])
transcript37 = get_transcript37(parsedLine[refSeqIndex])
transcript36 = get_transcript36(parsedLine[refSeqIndex])
# Normalize hgvs cdna string to fit what the counsyl hgvs parser determines to be the correct format
cdna_coord = str(
pyhgvs.format_hgvs_name(chrom38,
int(offset38),
ref38,
alt38,
genome38,
transcript38,
use_gene=False,
max_allele_length=100))
chrom38, offset38, ref38, alt38 = pyhgvs.parse_hgvs_name(
cdna_coord, genome38, get_transcript=get_transcript38)
chrom37, offset37, ref37, alt37 = pyhgvs.parse_hgvs_name(
cdna_coord, genome37, get_transcript=get_transcript37)
chrom36, offset36, ref36, alt36 = pyhgvs.parse_hgvs_name(
cdna_coord, genome36, get_transcript=get_transcript36)
# Generate transcript hgvs cdna synonym string
synonymString = []
if parsedLine[geneSymbolIndex] == 'BRCA1':
for transcriptName in refSeqBRCA1Transcripts:
transcript38 = get_transcript38(transcriptName)
cdna_synonym = str(
pyhgvs.format_hgvs_name(chrom38,
int(offset38),
ref38,
alt38,
genome38,
transcript38,
use_gene=False,
max_allele_length=100))
synonymString.append(cdna_synonym)
elif parsedLine[geneSymbolIndex] == 'BRCA2':
for transcriptName in refSeqBRCA2Transcripts:
transcript38 = get_transcript38(transcriptName)
cdna_synonym = str(
pyhgvs.format_hgvs_name(chrom38,
int(offset38),
ref38,
alt38,
genome38,
transcript38,
use_gene=False,
max_allele_length=100))
synonymString.append(cdna_synonym)
if calcProtein == True:
#print('oldHgvsGenomic38:', oldHgvsGenomic38)
#print('oldHgvsCDNA: ', oldHgvsCDNA)
#print('cdna: ', cdna_coord)
try:
var_c1 = hgvsparser.parse_hgvs_variant(cdna_coord)
protein_coord = variantmapper.c_to_p(var_c1)
except hgvs.exceptions.HGVSParseError as e:
print('hgvs.exceptions.HGVSParseError: ', e)
print(
'GRCh38 Genomic change: ',
'{0}:{1}:{2}>{3}'.format(chrom38, offset38, ref38, alt38))
print('')
#print('oldProtein: ', parsedLine[hgvsPIndex])
#print('protein:', protein_coord)
#print('')
# write new data into line
parsedLine[hgvsG36Index] = '{0}:{1}:{2}>{3}'.format(
chrom36, offset36, ref36, alt36)
parsedLine[hgvsG37Index] = '{0}:{1}:{2}>{3}'.format(
chrom37, offset37, ref37, alt37)
parsedLine[hgvsG38Index] = '{0}:{1}:{2}>{3}'.format(
chrom38, offset38, ref38, alt38)
parsedLine[hgvsCDNAIndex] = '{0}'.format(cdna_coord)
if calcProtein == True:
parsedLine[hgvsPIndex] = '{0}'.format(str(protein_coord))
parsedLine[synonymIndex] = ','.join(synonymString)
writeLine = '\t'.join(parsedLine) + '\n'
outputFile.writelines(writeLine)
hg18_fa.close()
hg19_fa.close()
hg38_fa.close()
refSeq18.close()
refSeq19.close()
refSeq38.close()
outputFile.close()