def get_transcript(name):
global REFGENE
if REFGENE is None:
sys.exit("No reference genome was provided. Try to locate hg38.BRCA.refGene.txt.")
with open(REFGENE) as infile:
TRANSCRIPTS = pyhgvs_utils.read_transcripts(infile)
return TRANSCRIPTS.get(name)
def parse_hgvs(hgvs_name, fasta, genes):
genome = Fasta(fasta, key_function=lambda x: 'chr{}'.format(x))
with open(genes) as infile:
transcripts = hgvs_utils.read_transcripts(infile)
def get_transcript(name):
return transcripts.get(name)
return hgvs.parse_hgvs_name(hgvs_name, genome, get_transcript=get_transcript)
def get_genome_coor(hgvs_c):
genome = SequenceFileDB('data/hg19.fa')
refGene = "/Users/Molly/Desktop/web-dev/hgvs_counsyl/hgvs/pyhgvs/data/genes.refGene"
with open(refGene) as infile:
transcripts = pyhgvs_utils.read_transcripts(infile)
def get_transcript(name):
return transcripts.get(name)
chrom, offset, ref, alt = pyhgvs.parse_hgvs_name(
hgvs_c, genome, get_transcript=get_transcript)
return chrom + ":" + str(offset) + ":" + ref + ">" + alt
def HGVS_to_GenomeCoor(HGVS):
"""use counsyl pyhgvs for this"""
genome = SequenceFileDB('../data/hg19.fa')
refGene = "../data/BRCA12.refGene.txt"
with open(refGene) as infile:
transcripts = pyhgvs_utils.read_transcripts(infile)
def get_transcript(name):
return transcripts.get(name)
chrom, offset, ref, alt = pyhgvs.parse_hgvs_name(
HGVS, genome, get_transcript=get_transcript)
genome_coordinate = chrom + ":" + str(offset) + ":" + ref + ">" + alt
return genome_coordinate
def get_genome_coor(hgvs_c):
genome = SequenceFileDB('data/hg19.fa')
refGene = "/Users/Molly/Desktop/web-dev/hgvs_counsyl/hgvs/pyhgvs/data/genes.refGene"
with open(refGene) as infile:
transcripts = pyhgvs_utils.read_transcripts(infile)
def get_transcript(name):
return transcripts.get(name)
chrom, offset, ref, alt = pyhgvs.parse_hgvs_name(
hgvs_c, genome, get_transcript=get_transcript)
return chrom + ":" + str(offset) + ":" + ref + ">" + alt
def __init__(self,ref_fa,ref_tr): #ref_tr is a file of transcripts formatted # like the refGene.txt file from ucsc #ref_fa is a reference file (genome) in fasta format #reference sequence self.ref_fa = ref_fa self.ref = SequenceFileDB(self.ref_fa) #refseq transcripts self.ref_tr = ref_tr with open(self.ref_tr) as infile: self.tr = util.read_transcripts(infile)
def calc_all(variants, priors, genome, transcripts, processes):
global brca1Transcript, brca2Transcript
inputData = csv.DictReader(variants, delimiter="\t")
fieldnames = inputData.fieldnames
newHeaders = open("headers.tsv", "r").read().split()
for header in newHeaders:
fieldnames.append(header)
outputData = csv.DictWriter(priors, delimiter="\t", lineterminator="\n", fieldnames=fieldnames)
outputData.writerow(dict((fn, fn) for fn in inputData.fieldnames))
# read RefSeq transcripts
transcripts = pyhgvs_utils.read_transcripts(transcripts)
brca1Transcript = transcripts.get(BRCA1_RefSeq)
brca2Transcript = transcripts.get(BRCA2_RefSeq)
if processes > 1:
# Create a pool of processes and calculate in parallel
click.echo("Processing using {} processes".format(processes), err=True)
pool = multiprocessing.Pool(processes)
try:
# Normal map has a bug if there is no timout that prevents Keyboard interrupts:
# https://stackoverflow.com/questions/1408356/keyboard-interrupts-with-pythons-multiprocessing-pool/1408476#1408476
# calc_one_partial = functools.partial(calc_one, brca1=brca1Transcript, brca2=brca2Transcript)
calculatedVariants = pool.map_async(calc_one, list(inputData)).get(99999999)
# Sort output as the order of p.map is not deterministic
outputData.writerows(sorted(
calculatedVariants,
key=lambda d: "{0}:g.{1}:{2}>{3}".format(d["Chr"], d["Pos"], d["Ref"], d["Alt"])))
except KeyboardInterrupt:
pool.terminate()
else:
outputData.writerows(sorted(
map(calc_one, inputData),
key=lambda d: "{0}:g.{1}:{2}>{3}".format(d["Chr"], d["Pos"], d["Ref"], d["Alt"])
))
def add_HGVS_c_counsyl(in_path, out_path):
# counsyl pyhgvs setups
with open('../data/BRCA12.refGene.txt') as infile:
transcripts_counsyl = pyhgvs_utils.read_transcripts(infile)
def get_transcript(name):
return transcripts_counsyl.get(name)
f_in = open(in_path, "r")
f_out = open(out_path, "w")
line_num = 0
unmatching_cases = 0
for line in f_in:
line_num += 1
print line_num
if line_num == 1:
f_out.write(line)
continue
items = line.strip().split("\t")
genome_coors = items[2].split(":")
chrom = genome_coors[0]
offset = int(genome_coors[1])
ref = genome_coors[2].split(">")[0]
alt = genome_coors[2].split(">")[1]
transcript_id = items[4]
transcript = get_transcript(transcript_id)
hgvs_name_with_transcript = pyhgvs.variant_to_hgvs_name(
chrom, offset, ref, alt, GENOME, transcript)
hgvs_c = hgvs_name_with_transcript.format(use_prefix=False,
use_gene=False)
if items[5] == "-":
items[5] = hgvs_c
elif items[5] == hgvs_c:
pass
else:
unmatching_cases += 1
new_line = "\t".join(items) + "\n"
f_out.write(new_line)
print "unmatching cases: ", unmatching_cases
def main():
options = parse_args()
inputFile = options.input
annotFile_path = options.inAnnot
vcfFile = options.out
genome_path = options.gpath
refseq_path = options.rpath
errorsFile = options.errors
source = options.source
with open(refseq_path) as infile:
transcripts = hgvs_utils.read_transcripts(infile)
genome = SequenceFileDB(genome_path)
def get_transcript(name):
return transcripts.get(name)
# open and store annotation fields in a dictionary
annotDict = defaultdict()
with open(annotFile_path) as inAnnotFile:
for line in inAnnotFile:
line = line.strip().split('\t')
annotDict[line[0]] = line[1]
# print header lines to vcf file
print('##fileformat=VCFv4.0', file=vcfFile)
if source == "exLOVD":
print('##source=exLOVD', file=vcfFile)
elif source == "LOVD":
print('##source=LOVD', file=vcfFile)
else:
raise ValueError('Source is %s, must be either LOVD or exLOVD' % (source))
print('##reference=GRCh37', file=vcfFile)
for annotation, description in annotDict.items():
print('##INFO=<ID={0},Number=.,Type=String,Description="{1}">'.format(annotation.replace(' ', '_'), description), file=vcfFile)
print('#CHROM\tPOS\tID\tREF\tALT\tQUAL\tFILTER\tINFO', file=vcfFile)
# extract INFO field column indicies for annotation terms
headerline = inputFile.readline().strip().replace(' ', '_').replace('"', '').split('\t')
fieldIdxDict = defaultdict()
for index, field in enumerate(headerline):
fieldIdxDict[field] = index
# extract info from each line of the flat file
for line in inputFile:
line = line.replace('"', '')
INFO_field = list()
parsedLine = line.strip().split('\t')
for field in headerline:
field_index = fieldIdxDict[field]
field_value = parsedLine[field_index]
field_value = normalize(field, field_value)
INFO_field.append('{0}={1}'.format(field, field_value))
# extract hgvs cDNA term for variant and cleanup formatting
# Sometimes dna_change is in the field cDNA, sometimes it's labeled dna_change.
if 'cDNA' in fieldIdxDict:
hgvsName = parsedLine[fieldIdxDict['cDNA']]
elif 'dna_change' in fieldIdxDict:
hgvsName = parsedLine[fieldIdxDict['dna_change']]
else:
sys.exit("ERROR: could not parse hgvs name.")
if hgvsName == '-':
print(parsedLine)
continue
queryHgvsName = hgvsName.rstrip().split(';')[0]
INFO_field_string = ';'.join(INFO_field)
try:
chrom, offset, ref, alt = hgvs.parse_hgvs_name(queryHgvsName, genome, get_transcript=get_transcript)
chrom = chrom.replace('chr', '')
print('{0}\t{1}\t{2}\t{3}\t{4}\t.\t.\t{5}'.format(chrom, offset, queryHgvsName, ref, alt, INFO_field_string), file=vcfFile)
except Exception as e:
print(str(e)+': could not parse hgvs field '+queryHgvsName, file=errorsFile)
#Load the data
df = pd.read_excel(io=path,
sheet_name=genename + '_' + variants,
engine='openpyxl')
data = []
for name in df['cDNA variant']:
data.append(name)
#read the genome
# the reference genome can be downloaded from: http://hgdownload.cse.ucsc.edu/goldenpath/hg19/bigZips/
genome = Fasta('references/hg19.fa')
# Read RefSeq transcripts into a python dict.
# The RefSeq transcripts can be downloaded from: https://github.com/counsyl/hgvs/blob/master/pyhgvs/data/genes.refGene
with open('references/genes.refGene') as infile:
transcripts = read_transcripts(infile)
# Provide a callback for fetching a transcript by its name.
def get_transcript(name):
return transcripts.get(name)
# Store the variant information in a list
vcf = []
for v in data:
chrom, offset, ref, alt = hgvs.parse_hgvs_name(
gene + ':' + v, genome, get_transcript=get_transcript)
vcf.append([chrom, offset, ref, alt])
# Define reference genome
def main(args):
options = parse_args()
exLOVDFile = options.inEXLOVD
annotFile_path = options.inAnnot
vcfFile = options.out
genome_path = options.gpath
refseq_path = options.rpath
errorsFile = options.errors
with open(refseq_path) as infile:
transcripts = hgvs_utils.read_transcripts(infile)
genome = SequenceFileDB(genome_path)
def get_transcript(name):
return transcripts.get(name)
# open and store annotation fields in a dictionary
annotDict = defaultdict()
with open(annotFile_path) as inAnnotFile:
for line in inAnnotFile:
line = line.strip().split('\t')
annotDict[line[0]] = line[1]
# print header lines to vcf file
print('##fileformat=VCFv4.0', file=vcfFile)
print('##source=exLOVD', file=vcfFile)
print('##reference=GRCh37', file=vcfFile)
for annotation, description in annotDict.items():
print('##INFO=<ID={0},Number=.,Type=String,Description="{1}">'.format(
annotation.replace(' ', '_'), description),
file=vcfFile)
print('#CHROM\tPOS\tID\tREF\tALT\tQUAL\tFILTER\tINFO', file=vcfFile)
# extract INFO field column indicies for annotation terms
headerline = exLOVDFile.readline().strip().replace(' ', '_').replace(
'"', '').split('\t')
fieldIdxDict = defaultdict()
for index, field in enumerate(headerline):
fieldIdxDict[field] = index
# extract info from each line of the bic flat file
for line in exLOVDFile:
line = line.replace('"', '')
INFO_field = list()
parsedLine = line.strip().split('\t')
for field in headerline:
field_index = fieldIdxDict[field]
field_value = parsedLine[field_index]
field_value = normalize(field, field_value)
INFO_field.append('{0}={1}'.format(field, field_value))
# extract hgvs cDNA term for variant and cleanup formatting
# Sometimes dna_change is in the field cDNA, sometimes it's labeled dna_change.
if 'cDNA' in fieldIdxDict:
hgvsName = parsedLine[fieldIdxDict['cDNA']]
elif 'dna_change' in fieldIdxDict:
hgvsName = parsedLine[fieldIdxDict['dna_change']]
else:
sys.exit("ERROR: could not parse hgvs name.")
if hgvsName == '-':
print(parsedLine)
continue
queryHgvsName = hgvsName.rstrip().split(';')[0]
INFO_field_string = ';'.join(INFO_field)
try:
chrom, offset, ref, alt = hgvs.parse_hgvs_name(
queryHgvsName, genome, get_transcript=get_transcript)
chrom = chrom.replace('chr', '')
print('{0}\t{1}\t{2}\t{3}\t{4}\t.\t.\t{5}'.format(
chrom, offset, queryHgvsName, ref, alt, INFO_field_string),
file=vcfFile)
except Exception as e:
print(str(e) + ': could not parse hgvs field ' + queryHgvsName,
file=errorsFile)
def get_transcript(name):
REFGENE = "../refgene38_brca.txt"
with open(REFGENE) as infile:
TRANSCRIPTS = pyhgvs_utils.read_transcripts(infile)
return TRANSCRIPTS.get(name)
def main():
options = parse_args()
inputFile = options.input
annotFile_path = options.inAnnot
vcfFile = options.out
genome_path = options.gpath
refseq_path = options.rpath
source = options.source
logfile = options.logfile
if options.verbose:
logging_level = logging.DEBUG
else:
logging_level = logging.CRITICAL
logging.basicConfig(filename=logfile, filemode="w", level=logging_level)
with open(refseq_path) as infile:
transcripts = hgvs_utils.read_transcripts(infile)
genome = SequenceFileDB(genome_path)
def get_transcript(name):
return transcripts.get(name)
# open and store annotation fields in a dictionary
annotDict = defaultdict()
with open(annotFile_path) as inAnnotFile:
for line in inAnnotFile:
line = line.strip().split('\t')
annotDict[line[0]] = line[1]
# print header lines to vcf file
print('##fileformat=VCFv4.0', file=vcfFile)
print('##source={0}'.format(source), file=vcfFile)
print('##reference=GRCh37', file=vcfFile)
for annotation, description in annotDict.items():
print('##INFO=<ID={0},Number=.,Type=String,Description="{1}">'.format(
annotation.replace(' ', '_'), description),
file=vcfFile)
print('#CHROM\tPOS\tID\tREF\tALT\tQUAL\tFILTER\tINFO', file=vcfFile)
# extract INFO field column indicies for annotation terms
headerline = inputFile.readline().strip().replace(' ', '_').replace(
'"', '').split('\t')
fieldIdxDict = defaultdict()
for index, field in enumerate(headerline):
fieldIdxDict[field] = index
# extract info from each line of the flat file
for line in inputFile:
line = line.replace('"', '')
INFO_field = list()
parsedLine = line.strip().split('\t')
for field in headerline:
field_index = fieldIdxDict[field]
field_value = parsedLine[field_index]
field_value = normalize(field, field_value)
INFO_field.append('{0}={1}'.format(field, field_value))
# extract hgvs cDNA term for variant and cleanup formatting
hgvsName = parsedLine[fieldIdxDict['hgvs_nucleotide']]
if hgvsName == '-':
logging.debug("hgvs name == '-' for line: %s", parsedLine)
continue
gene_symbol = parsedLine[fieldIdxDict['gene_symbol']].lower()
if gene_symbol == 'brca1':
transcript = 'NM_007294.3'
elif gene_symbol == 'brca2':
transcript = 'NM_000059.3'
else:
logging.debug("improper gene symbol: %s", gene_symbol)
continue
queryHgvsName = transcript + ':' + hgvsName.rstrip().split(';')[0]
INFO_field_string = ';'.join(INFO_field)
try:
chrom, offset, ref, alt = hgvs.parse_hgvs_name(
queryHgvsName, genome, get_transcript=get_transcript)
chrom = chrom.replace('chr', '')
print('{0}\t{1}\t{2}\t{3}\t{4}\t.\t.\t{5}'.format(
chrom, offset, queryHgvsName, ref, alt, INFO_field_string),
file=vcfFile)
except Exception as e:
logging.debug("could not parse hgvs field: %s", queryHgvsName)
config = configparser.ConfigParser()
config.read(BinPath + '/config.ini')
omim_dict = read_morbidmap(BinPath + '/' + config['DEFAULT']['morbidmap'])
pathogenic_dict, pathogenic_dict2 = read_pathogenic_site(
BinPath + '/' + config['DEFAULT']['pathogenic_ref'])
ba1_exception = read_ba1_exception(BinPath + '/' +
config['DEFAULT']['ba1_exception'])
pvs1_levels = read_pvs1_levels(BinPath + '/' + config['DEFAULT']['pvs1levels'])
domain_bed = create_bed_dict(BinPath + '/' + config['DEFAULT']['domain'])
hotspot_bed = create_bed_dict(BinPath + '/' + config['DEFAULT']['hotspot'])
curated_region = create_bed_dict(BinPath + '/' +
config['DEFAULT']['curated_region'])
exon_lof_frequent = create_bed_dict(BinPath + '/' +
config['DEFAULT']['exon_lof_frequent'])
genome = Fasta(BinPath + '/' + config['DEFAULT']['ref'])
with open(BinPath + '/' + config['DEFAULT']['trans']) as gpefile:
transcripts = read_transcripts(gpefile)
gene_trans = {}
trans_gene = {}
with open(BinPath + '/' + config['DEFAULT']['trans']) as f:
for line in f:
record = line.strip().split("\t")
gene = record[12]
trans = record[1]
gene_trans[gene] = trans
trans_gene[trans] = gene
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()
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 get_transcript(name): REFGENE = "../refgene38_brca.txt" with open(REFGENE) as infile: TRANSCRIPTS = pyhgvs_utils.read_transcripts(infile) return TRANSCRIPTS.get(name)
def main(args):
options = parse_args()
hdp = hgvs.dataproviders.uta.connect()
am38 = hgvs.assemblymapper.AssemblyMapper(hdp, assembly_name='GRCh38')
hn = hgvs.normalizer.Normalizer(hdp)
hp = hgvs.parser.Parser()
# Read genome sequence using pyfaidx
genome = Fasta(options.refFASTA)
# Read RefSeq transcripts into a python dict.
with open(options.refSEQ) as infile:
transcripts = pyhgvs_utils.read_transcripts(infile)
# Provide a callback for fetching a transcript by its name.
def get_transcript(name):
return transcripts.get(name)
babelfish38 = Babelfish(hdp, assembly_name="GRCh38")
## extract base variant representation
with open(options.inVCF, 'rb') as in_vcf, open(options.outVCF,
'w') as out_vcf:
vcf_reader = vcf.Reader(in_vcf)
vcf_writer = vcf.Writer(out_vcf, vcf_reader)
for record in vcf_reader:
# Convert variants for indel HGVS representation
chrom, offset, ref, alt = (str(record.CHROM), record.POS,
str(record.REF), str(record.ALT[0]))
print('chrom: {}, offset: {}, ref: {}, alt: {}'.format(
chrom, offset, ref, alt))
if 'chr13' in record.CHROM:
transcript_id = "NM_000059.3"
elif 'chr17' in record.CHROM:
transcript_id = "NM_007294.4"
transcript = get_transcript(transcript_id)
try:
hgvs_name = pyhgvs.format_hgvs_name(chrom,
offset,
ref,
alt,
genome,
transcript,
use_gene=False,
max_allele_length=50000)
hgvs_c = hp.parse_hgvs_variant(hgvs_name)
if len(ref) == len(alt) and len(ref) == 1:
# Variant is a SNP, normalize using hgvs Normalizer function
if 'chr17' in record.CHROM: hgvs_c.ac = 'NM_007294.3'
norm_hgvs_c = hn.normalize(hgvs_c)
if 'chr17' in record.CHROM: norm_hgvs_c.ac = 'NM_007294.4'
chrom, offset, ref, alt = pyhgvs.parse_hgvs_name(
str(norm_hgvs_c),
genome,
normalize=False,
get_transcript=get_transcript)
else:
# Variant is an INDEL, normalize using hgvs babelfish38.hgvs_to_vcf function
if 'chr17' in record.CHROM: hgvs_c.ac = 'NM_007294.3'
hgvs_g = am38.c_to_g(hgvs_c)
vcf_values = babelfish38.hgvs_to_vcf(hgvs_g)
chrom, offset, ref, alt = 'chr{}'.format(
vcf_values[0]
), vcf_values[1], vcf_values[2], vcf_values[3]
except hgvs.exceptions.HGVSUnsupportedOperationError as e:
print(
'hgvs.exceptions.HGVSUnsupportedOperationError: {}'.format(
e))
except hgvs.exceptions.HGVSInvalidIntervalError as e:
print('hgvs.exceptions.HGVSInvalidIntervalError: {}'.format(e))
except hgvs.exceptions.HGVSInvalidVariantError as e:
print('hgvs.exceptions.HGVSInvalidVariantError: {}'.format(e))
except AttributeError as e:
print('AttributeError: {}'.format(e))
except KeyError as e:
print('KeyError: {}'.format(e))
# Update and write the new normalized record
record.POS = offset
record.REF = ref
record.ALT = [alt]
vcf_writer.write_record(record)
def get_transcript(name):
REFGENE = "../resources/refseq/hg38.BRCA.refGene.txt"
with open(REFGENE) as infile:
TRANSCRIPTS = pyhgvs_utils.read_transcripts(infile)
return TRANSCRIPTS.get(name)
import toolshed as ts
from collections import defaultdict
import pyhgvs as hgvs
from pygr.seqdb import SequenceFileDB
from pyhgvs.utils import read_transcripts
import sys
var=sys.argv[1] #'ogfiles/fetalgenomevariants.txt'
trans=sys.argv[2] #'ogfiles/fetaltranscripts.txt'
f=open(sys.argv[3],'w') #'fetalvariants.vcf'
infile='Homo_sapiens.GRCh37.75.genePred' #obtained by running gtfToGenePred UCSC binary on the GTF of GRCh37 from ENSEMBL. added 0 as placeholder undocumented id field to the front of each line using sed
with open(infile) as reffile:
transcripts=read_transcripts(reffile)
def get_transcript(name):
return transcripts.get(name)
def translate(variant,transcripts,get_transcript):
genome = SequenceFileDB('hg19.fa') #pip install bsddb3 is required
try:
chrom, offset, ref, alt = hgvs.parse_hgvs_name(variant, genome, get_transcript=get_transcript)
except:
return 1
return chrom, offset, ref, alt
def readgenes(trans):
genes=defaultdict(str)
for fields in (x.rstrip('\r\n').split("\t") for x in ts.nopen(trans)):
gene=fields[0]; transcript=fields[1]
genes[gene]=transcript
return genes
def main():
options = parse_args()
inputFile = options.input
annotFile_path = options.inAnnot
vcfFile = options.out
genome_path = options.gpath
refseq_path = options.rpath
source = options.source
logfile = options.logfile
if options.verbose:
logging_level = logging.DEBUG
else:
logging_level = logging.CRITICAL
logging.basicConfig(filename=logfile, filemode="w", level=logging_level)
with open(refseq_path) as infile:
transcripts = hgvs_utils.read_transcripts(infile)
genome = SequenceFileDB(genome_path)
def get_transcript(name):
return transcripts.get(name)
# open and store annotation fields in a dictionary
annotDict = defaultdict()
with open(annotFile_path) as inAnnotFile:
for line in inAnnotFile:
line = line.strip().split('\t')
annotDict[line[0]] = line[1]
# print header lines to vcf file
print('##fileformat=VCFv4.0', file=vcfFile)
print('##source={0}'.format(source), file=vcfFile)
print('##reference=GRCh37', file=vcfFile)
for annotation, description in annotDict.items():
print('##INFO=<ID={0},Number=.,Type=String,Description="{1}">'.format(annotation.replace(' ', '_'), description), file=vcfFile)
print('#CHROM\tPOS\tID\tREF\tALT\tQUAL\tFILTER\tINFO', file=vcfFile)
# extract INFO field column indicies for annotation terms
headerline = inputFile.readline().strip().replace(' ', '_').replace('"', '').split('\t')
fieldIdxDict = defaultdict()
for index, field in enumerate(headerline):
fieldIdxDict[field] = index
# extract info from each line of the flat file
for line in inputFile:
line = line.replace('"', '')
INFO_field = list()
parsedLine = line.strip().split('\t')
for field in headerline:
field_index = fieldIdxDict[field]
field_value = parsedLine[field_index]
field_value = normalize(field, field_value)
INFO_field.append('{0}={1}'.format(field, field_value))
# extract hgvs cDNA term for variant and cleanup formatting
hgvsName = parsedLine[fieldIdxDict['hgvs_nucleotide']]
if hgvsName == '-':
logging.debug("hgvs name == '-' for line: %s", parsedLine)
continue
gene_symbol = parsedLine[fieldIdxDict['gene_symbol']].lower()
if gene_symbol == 'brca1':
transcript = 'NM_007294.3'
elif gene_symbol == 'brca2':
transcript = 'NM_000059.3'
else:
logging.debug("improper gene symbol: %s", gene_symbol)
continue
queryHgvsName = transcript + ':' + hgvsName.rstrip().split(';')[0]
INFO_field_string = ';'.join(INFO_field)
try:
chrom, offset, ref, alt = hgvs.parse_hgvs_name(queryHgvsName, genome, get_transcript=get_transcript)
chrom = chrom.replace('chr', '')
print('{0}\t{1}\t{2}\t{3}\t{4}\t.\t.\t{5}'.format(chrom, offset, queryHgvsName, ref, alt, INFO_field_string), file=vcfFile)
except Exception as e:
logging.debug("could not parse hgvs field: %s", queryHgvsName)
import pyhgvs
import pyhgvs.utils as pyhgvs_utils
from pygr.seqdb import SequenceFileDB
import sys
import os
import json
import string_comp
ORIGINAL_FILE = "../BRCA_selectedLabs_only/ClinVarBRCA.selectedLabsOnly.txt"
ERROR = "../BRCA_selectedLabs_only/BRCA.wrong_genome_Coor"
FILE = "../BRCA_selectedLabs_only/BRCA.pre-processed"
GENOME = SequenceFileDB("../reference_files/hg19.fa")
with open('../reference_files/genes.refGene.BRCA.txt') as infile:
transcripts = pyhgvs_utils.read_transcripts(infile)
def get_transcript(name):
return transcripts.get(name)
def main():
check_HGVS_conversion_error()
#HGVS_conversion()
def HGVS_to_genome_coor(HGVS):
try:
chrm, pos, ref, alt = pyhgvs.parse_hgvs_name(
HGVS, GENOME, get_transcript=get_transcript)
chrm = chrm[3:]
pos = str(pos)
('NM_000352.3', 'c', '>', CDNACoord(215, -10), CDNACoord(215, -10), 'A', 'G')
"""
from __future__ import print_function
from __future__ import unicode_literals
import pyhgvs as hgvs
import pyhgvs.utils as hgvs_utils
from pyfaidx import Fasta
# Read genome sequence using pyfaidx.
genome = Genome('/tmp/hg19.fa')
# Read RefSeq transcripts into a python dict.
with open('pyhgvs/data/genes.refGene') as infile:
transcripts = hgvs_utils.read_transcripts(infile)
# Provide a callback for fetching a transcript by its name.
def get_transcript(name):
return transcripts.get(name)
# Parse the HGVS name into genomic coordinates and alleles.
chrom, offset, ref, alt = hgvs.parse_hgvs_name('NM_000352.3:c.215A>G',
genome,
get_transcript=get_transcript)
print(chrom, offset, ref, alt)
# Returns variant in VCF style: ('chr11', 17496508, 'T', 'C')
# Notice that since the transcript is on the negative strand, the alleles
# are reverse complemented during conversion.
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)
hgvs_parser = hgvs.parser.Parser()
hgvs_dp = hgvs.dataproviders.uta.connect()
hgvs_norm = hgvs.normalizer.Normalizer(hgvs_dp)
hgvs_am = hgvs.assemblymapper.AssemblyMapper(hgvs_dp, assembly_name='GRCh38')
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
cdna_coord_LOVD = cdna_coord_LOVD.strip()
# 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
try:
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)
except Exception as e:
print('parse error: {}'.format(cdna_coord_LOVD))
print(e)
protein_coord = None
if calcProtein:
try:
genomic_change = '{0}:g.{1}:{2}>{3}'.format(chrom38, offset38, ref38, alt38)
var_c1 = hgvs_parser.parse_hgvs_variant(cdna_coord)
var_c1_norm = hgvs_norm.normalize(var_c1) # doing normalization explicitly to get a useful error message
protein_coord = hgvs_am.c_to_p(var_c1_norm)
except Exception as e:
template = "An error of type {0} occured. Arguments:{1!r}"
error_name = type(e).__name__
message = template.format(error_name, e.args)
logging.error(message)
logging.error('Proposed GRCh38 Genomic change for error: %s', genomic_change)
logging.error(line)
# Exceptions related to invalid data
data_errors = set(['HGVSParseError', 'HGVSError', 'HGVSInvalidVariantError', 'HGVSUnsupportedOperationError'])
if error_name not in data_errors:
# output some more if exception doesn't seem to be related to invalid data
logging.error("Non data error raised")
logging.exception(message)
if error_name == "DatabaseError":
# Aborting, as it is a transient error in principle, i.e. in one run we might be able to obtain a protein change, in another not, messing up the data diffs
raise EnvironmentError("Issue with UTA database. Aborting")
# 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()