def check_HGVS_conversion_error():
df = pd.read_csv(ORIGINAL_FILE, sep="\t")
n_diff = 0
n_real_diff = 0
for row in df.iterrows():
original_genome_coor = [row[1].Chrom, row[1].Pos,
row[1].Ref.replace("-", ""),
row[1].Alt.replace("-", "")]
try:
chrm, pos, ref, alt = pyhgvs.parse_hgvs_name(
row[1].HGVS, GENOME, get_transcript=get_transcript)
chrm = chrm[3:]
pos = str(pos)
converted_genome_coor = [chrm, pos, ref, alt]
except (pyhgvs.InvalidHGVSName, NotImplementedError, ValueError, AssertionError):
converted_genome_coor = ["NA"]
if converted_genome_coor == original_genome_coor:
continue
elif "NA" in converted_genome_coor or "None" in original_genome_coor:
continue
else:
n_diff += 1
if not string_comp.variant_equal(original_genome_coor, converted_genome_coor):
n_real_diff += 1
print "total number of rows in file: ", len(df)
print "number of mismatching result by direct comparison: ", n_diff
print "number of mismatching result by string comparison: ", n_real_diff
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 hgvs_to_vcf(self, hgvs_variant):
"""
Converts a single variant provided in HGVS notation to genomic coordinate notation.
See Counsyl's HGVS library for more information on acceptable input formats: https://github.com/counsyl/hgvs.
Args:
hgvs_variant (str): HGVS description of variant, such as NM_001100.3:c.137T>C.
The portion prior to the colon is the refseqID used as the reference for the variant The portion after
the colon is an HGVS-style description of the mutation (a SNP from T to C at location 137 in the example above.
Returns:
tuple of str: (chromosome_number, coordinate, ref, alt) in that order denoting the VCF notation of the variant
"""
# Library requires string not unicode, ensure format is correct
hgvs_variant = str(hgvs_variant)
chromosome_number, coordinate, ref, alt = pyhgvs.parse_hgvs_name(
hgvs_variant, self.genome, get_transcript=self._get_transcript)
chromosome_number = re.match('chr(.+)', chromosome_number).group(1)
coordinate = str(coordinate)
return chromosome_number, coordinate, ref, alt
def to_chrom_coordinate(self, cDNA):
try:
chrom, offset, ref, alt = pyhgvs.parse_hgvs_name(cDNA, self.genome, \
get_transcript = self.get_transcript)
return chrom, offset, ref, alt
except:
print "[%s] cannot be coverted to chromosome coordinate" % cDNA
return None, None, None, None
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 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)
ref = ref.replace("-", "")
alt = alt.replace("-", "")
genome_coor = "_".join([chrm, pos, ref, alt])
except(pyhgvs.InvalidHGVSName, NotImplementedError, ValueError, AssertionError):
genome_coor = "not translated"
return genome_coor
def egl_variants(filename):
print('\
\n----------------------------------\
\n\
\n EGL Database \
\n\
\n----------------------------------')
# Load file
egl = pd.read_csv(filename, usecols=range(0, 9), header=None)
# Rename columns accordingly
egl.columns = [
'ID', 'gene', '', 'exon', 'hgvs', 'protein_change', 'sig',
'last reviewed', 'alias listing'
]
# Filter LAMA2
lama2_egl = egl[egl.iloc[:, 1].str.match(r'LAMA2')]
lama2_egl = lama2_egl.loc[:, ['ID', 'hgvs', 'sig']]
lama2_egl = lama2_egl.replace({'sig': dsig})
lama2_egl['ID'] = 'egl_' + lama2_egl['ID'].astype(str)
genome = Fasta('reference/chr6.fa')
# Converting HGVS to VCF genomic coordinates
count = 0
transcript = get_transcript('NM_000426')
for i in lama2_egl.index:
try:
#CHROM, POS, REF, ALT = pyhgvs.parse_hgvs_name(lama2_egl.loc[i,'hgvs'], genome, get_transcript=get_transcript)
CHROM, POS, REF, ALT = pyhgvs.parse_hgvs_name(
lama2_egl.loc[i, 'hgvs'], genome, transcript)
lama2_egl.loc[i, "CHROM"] = 6
lama2_egl.loc[i, "POS"] = POS
lama2_egl.loc[i, "REF"] = REF
lama2_egl.loc[i, "ALT"] = ALT
except:
print('\nException:', lama2_egl.loc[i, 'hgvs'])
count += 1
print('Number of exceptions:', count)
pass
lama2_egl = lama2_egl.astype({'CHROM': 'int32', 'POS': 'int32'})
# Pathogenic variants
#pathogenic_egl = lama2_egl[lama2_egl.sig.str.match(r'[Pp]athogenic')]
#return [lama2_egl, pathogenic_egl]
return lama2_egl
def test_pyhgvs_cdna_coordinate_correct(self):
for i in self.data:
pyhgvs_coord = i['pyhgvs_Genomic_Coordinate_38']
pyhgvs_cDNA = i['pyhgvs_cDNA']
genome = SequenceFileDB('../reference_genome/hg38/hg38.fa')
def get_transcript(name):
REFGENE = "../refgene38_brca.txt"
with open(REFGENE) as infile:
TRANSCRIPTS = pyhgvs_utils.read_transcripts(infile)
return TRANSCRIPTS.get(name)
chrom, offset, ref, alt = pyhgvs.parse_hgvs_name(pyhgvs_cDNA, genome,
get_transcript=get_transcript)
test_coord = chrom + ":" + "g." + str(offset) + ":" + ref + ">" + alt
self.assertEqual(pyhgvs_coord, test_coord)
def test_pyhgvs_cdna_coordinate_correct(self):
for i in self.data:
pyhgvs_coord = i['pyhgvs_Genomic_Coordinate_38']
pyhgvs_cDNA = i['pyhgvs_cDNA']
genome = SequenceFileDB('../reference_genome/hg38/hg38.fa')
def get_transcript(name):
REFGENE = "../refgene38_brca.txt"
with open(REFGENE) as infile:
TRANSCRIPTS = pyhgvs_utils.read_transcripts(infile)
return TRANSCRIPTS.get(name)
chrom, offset, ref, alt = pyhgvs.parse_hgvs_name(
pyhgvs_cDNA, genome, get_transcript=get_transcript)
test_coord = chrom + ":" + "g." + str(
offset) + ":" + ref + ">" + alt
self.assertEqual(pyhgvs_coord, test_coord)
def check_hgvs(hgvs, submitterteam, submitter, file):
try:
chrom, offset, ref, alt = pyhgvs.parse_hgvs_name(
str(hgvs), genome, get_transcript=get_transcript)
except ValueError, e: # 'falsche' HGVS-Codes überspringen und anzeigen
print e, hgvs
if file not in overview[submitterteam]['incorrect JSONs'].keys():
overview[submitterteam]['incorrect JSONs'][file] = {}
if 'falscher HGVS-Code' not in overview[submitterteam][
'incorrect JSONs'][file]:
overview[submitterteam]['incorrect JSONs'][file][
'falscher HGVS-Code'] = {}
overview[submitterteam]['incorrect JSONs'][file][
'submitter'] = submitter
overview[submitterteam]['incorrect JSONs'][file][
'falscher HGVS-Code'][hgvs] = str(e)
else:
overview[submitterteam]['incorrect JSONs'][file][
'falscher HGVS-Code'][hgvs] = str(e)
def hgvs_to_vcf(self, hgvs_variant):
"""
Converts a single variant provided in HGVS notation to genomic coordinate notation.
See Counsyl's HGVS library for more information on acceptable input formats: https://github.com/counsyl/hgvs.
Args:
hgvs_variant (str): HGVS description of variant, such as NM_001100.3:c.137T>C.
The portion prior to the colon is the refseqID used as the reference for the variant The portion after
the colon is an HGVS-style description of the mutation (a SNP from T to C at location 137 in the example above.
Returns:
tuple of str: (chromosome_number, coordinate, ref, alt) in that order denoting the VCF notation of the variant
"""
# Library requires string not unicode, ensure format is correct
hgvs_variant = str(hgvs_variant)
chromosome_number, coordinate, ref, alt = pyhgvs.parse_hgvs_name(hgvs_variant, self.genome, get_transcript=self._get_transcript)
chromosome_number = re.match('chr(.+)', chromosome_number).group(1)
coordinate = str(coordinate)
return chromosome_number, coordinate, ref, alt
else:
print 'keine Mutationen: ', file
continue
genotype = mutation['Test Information']['Genotype']
if genotype == 'Hemizygous':
genotype = '1'
elif genotype == 'Homozygous':
genotype = '1/1'
elif genotype == 'Heterozygous' or genotype == 'Compound Heterozygous':
genotype = '0/1'
else:
genotype = './1'
for hgvscode in hgvslist:
try:
chrom, offset, ref, alt = pyhgvs.parse_hgvs_name(
str(hgvscode),
genome,
get_transcript=get_transcript)
if hgvscode in multivcf['NM'].tolist():
index = multivcf['NM'].tolist().index(hgvscode)
multivcf.set_value(index, caseID, genotype)
if caseID not in jsonlist2:
jsonlist2.append(caseID)
else:
chromo = chrom.split('chr')[1]
multivcf.set_value(x, '#CHROM', str(chromo))
try:
multivcf.set_value(x, 'sort', int(chromo))
except ValueError, e:
multivcf.set_value(x, 'sort', 30)
multivcf.set_value(x, 'NM', hgvscode)
multivcf.set_value(x, 'POS', offset)
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()
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 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)
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()
#!/usr/bin/env python
import argparse
import pyhgvs as hgvs
import hgvs.utils as hgvs_utils
from pygr.seqdb import SequenceFileDB
if __name__ == "__main__":
parser = argparse.ArgumentParser()
parser.add_argument('-i', '--infile', help="Input file")
parser.add_argument('-g', '--genome', help="Genome reference FASTA")
parser.add_argument('-r', '--refgene', help="RefGene Transcripts")
parser.add_argument('-o', '--outfile', help="Output file name")
args = parser.parse_args()
genome = SequenceFileDB(args.genome)
with open(args.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)
chrom, offset, ref, alt = hgvs.parse_hgvs_name('NM_000352.3:c.215A>G', genome, get_transcript=get_transcript)
def lovd_variants(filename):
print('\
\n--------------------------------------\
\n\
\n LOVD Database \
\n\
\n--------------------------------------')
# Load the file
lovd_ori = pd.read_csv(filename,
header=0,
quotechar='"',
doublequote=True,
sep="\t",
skiprows=4110,
nrows=2055)
# Cleaning up the names of columns
lovd_ori.columns = lovd_ori.columns.str.replace("['\{{','\}}']", '')
lovd_ori.columns = lovd_ori.columns.str.replace("VariantOnGenome/", '')
lovd_ori = lovd_ori.drop_duplicates(subset='DNA/hg38')
lovd_ori['DNA/hg38'] = "NC_000006.12:" + lovd_ori['DNA/hg38'].astype(str)
lovd = lovd_ori.rename(columns={
'DBID': 'ID',
'DNA/hg38': 'hgvs',
'ClinicalClassification': 'sig'
})
# Selecting important columns
lovd = lovd.loc[:, ['ID', 'hgvs', 'sig', 'type']]
lovd = lovd.replace({'type': dtype})
genome = Fasta('reference/NC_000006.fa')
# Converting HGVS to VCF genomic coordinates
count = 0
for i in lovd.index:
try:
#print('Trying %d\n' %(lovd.loc[i,'hgvs']))
CHROM, POS, REF, ALT = pyhgvs.parse_hgvs_name(
lovd.loc[i, 'hgvs'], genome)
lovd.loc[i, 'CHROM'] = 6
lovd.loc[i, "POS"] = POS
lovd.loc[i, "REF"] = REF
lovd.loc[i, "ALT"] = ALT
except:
print('\nException:', lovd.loc[i, 'hgvs'])
count += 1
print('Number of exceptions:', count)
pass
# Filtering pathogenic variants
#pathogenic_lovd = lovd[lovd.sig.str.match(r'.*[Pp]athogenic.*', na=False)]
#return [lovd, pathogenic_lovd]
lovd.dropna(inplace=True)
lovd = lovd.astype({'CHROM': 'int32', 'POS': 'int32'})
#print(lovd)
return lovd
import pyximport
pyximport.install()
import pyhgvs as hgvs
import pyhgvs.utils as hgvs_utils
import pygr
from pygr.seqdb import SequenceFileDB
from optparse import OptionParser
parser = OptionParser()
parser.add_option("-i", "--input_string", dest="instr", help="format NM_007294.3:c.8T>G", default="")
(options, args) = parser.parse_args()
instr = options.instr
# Read genome sequence using pygr.
fn_hg19='/raw/human_sequence_reference/UCSC/hg19.fa'
fn_refgene='/home/david/software/hgvs-master/pyhgvs/data/genes.refGene'
genome = SequenceFileDB(fn_hg19)
def get_transcript(name):
return transcripts.get(name)
with open(fn_refgene) as infile:
transcripts = hgvs_utils.read_transcripts(infile)
out = hgvs.parse_hgvs_name(instr,genome,get_transcript=get_transcript)
print( "\t".join( [str(x) for x in out] ) )
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)
# 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
ref_genome = [
'reference=GRCh37/hg19', 'contig=<ID=1,length=249250621>',
'contig=<ID=2,length=243199373>', 'contig=<ID=3,length=198022430>',
'contig=<ID=4,length=191154276>', 'contig=<ID=5,length=180915260>',
'contig=<ID=6,length=171115067>', 'contig=<ID=7,length=159138663>',
'contig=<ID=8,length=146364022>', 'contig=<ID=9,length=141213431>',
'contig=<ID=10,length=135534747>', 'contig=<ID=11,length=135006516>',
'contig=<ID=12,length=133851895>', 'contig=<ID=13,length=115169878.',
'contig=<ID=14,length=107349540>', 'contig=<ID=15,length=102531392>',
'contig=<ID=16,length=90354753>', 'contig=<ID=17,length=81195210>',
'contig=<ID=18,length=78077248>', 'contig=<ID=19,length=59128983>',
'contig=<ID=20,length=63025520>', 'contig=<ID=21,length=48129895>',
if '_' in cdot.split(':')[1]:
return fix_alt_range(chrom, offset, ref, cdot)
return fix_alt_single(chrom, offset, ref, cdot)
dat_file = '../../data/raw/UC_all_panel_variants_01_20_2016.xlsx'
df_pre = pandas.read_excel(dat_file)
crit = df_pre.apply(
lambda row: 'ins' in row['Alt'] and row['Gene Symbol'] in FOCUS_GENES,
axis=1)
df = df_pre[crit]
df.loc[:, 'id'] = df.apply(
lambda row: row['Transcript'] + ':' + row['Pos'] + row['Alt'], axis=1)
vals = list(df[['id', 'Classification']].values)
for val in vals:
# print(val)
v, c = val
print(v, c)
chrom, offset, ref, alt = hgvs.parse_hgvs_name(
str(v), genome, get_transcript=get_transcript)
print(v, c, chrom, offset, ref, alt)
# h = str(hgvs.HGVSName(v)).split("'")[1]
# print(v)
# try:
# except:
# print('FAIL', v, c)
# next convert to g. https://github.com/counsyl/hgvs
# need hg19
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)
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)
from pygr.seqdb import SequenceFileDB
if __name__ == "__main__":
parser = argparse.ArgumentParser()
parser.add_argument('-i', '--infile', help="Input file")
parser.add_argument('-g', '--genome', help="Genome reference FASTA")
parser.add_argument('-r', '--refgene', help="RefGene Transcripts")
parser.add_argument('-o', '--outfile', help="Output file name")
args = parser.parse_args()
genome = SequenceFileDB(args.genome)
with open(args.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)
with open(args.infile, 'r') as infile:
with open(args.outfile, 'w') as outfile:
reader = csv.reader(infile, dialect='excel-tab')
header = reader.next()
for row in reader:
print row[1]
chrom, offset, ref, alt = hgvs.parse_hgvs_name(row[1], genome, get_transcript=get_transcript)
outfile.write("{}\t{}\t{}\t{}\n".format(chrom, offset, ref, alt))
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)
def convert_HGVS(hgvs_c, GENOME):
chrom, offset, ref, alt = pyhgvs.parse_hgvs_name(
hgvs_c, GENOME, get_transcript=get_transcript)
genome_coor = chrom + ":" + str(offset) + ":" + ref + ">" + alt
HGVS_p = HGVS_cDNA_to_protein(hgvs_c)
return genome_coor, HGVS_p
# 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.
# Format an HGVS name.
chrom, offset, ref, alt = ('chr11', 17496508, 'T', 'C')
transcript = get_transcript('NM_000352.3')
hgvs_name = hgvs.format_hgvs_name(chrom, offset, ref, alt, genome, transcript)
print(hgvs_name)
# Returns 'NM_000352.3(ABCC8):c.215A>G'
hgvs_name = hgvs.HGVSName('NM_000352.3:c.215-10A>G')
# fields of the HGVS name are available as attributes:
from pygr.seqdb import SequenceFileDB
if __name__ == "__main__":
parser = argparse.ArgumentParser()
parser.add_argument('-i', '--infile', help="Input file")
parser.add_argument('-g', '--genome', help="Genome reference FASTA")
parser.add_argument('-r', '--refgene', help="RefGene Transcripts")
parser.add_argument('-o', '--outfile', help="Output file name")
args = parser.parse_args()
genome = SequenceFileDB(args.genome)
with open(args.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)
with open(args.infile, 'r') as infile:
with open(args.outfile, 'w') as outfile:
reader = csv.reader(infile, dialect='excel-tab')
header = reader.next()
for row in reader:
print row[1]
chrom, offset, ref, alt = hgvs.parse_hgvs_name(
row[1], genome, get_transcript=get_transcript)
outfile.write("{}\t{}\t{}\t{}\n".format(
chrom, offset, ref, alt))
