def parse_hgvs_muts_file(hgvs_muts_file, raise_exception=True):
hgvs_muts = []
for m in [l.rstrip('\n') for l in open(hgvs_muts_file).readlines()]:
if raise_exception:
hgvs_mut = hgvs.HGVSName(m)
hgvs_muts += [hgvs_mut]
else:
try:
hgvs_mut = hgvs.HGVSName(m)
hgvs_muts += [hgvs_mut]
except hgvs.InvalidHGVSName:
sys.stderr.write("Invalid HGVS found: {}\n".format(m))
pass
return hgvs_muts
def convertGenomicPosToTranscriptPos(genomicPos, chrom, genome, transcript):
"""
Given a genomic position, chrom (in format "chrN"), genome (SequenceFileDB for genome),
and transcript (pyhgvs transcript object):
Returns a string of the transcript position at the given genomic position
"""
# use "T" and "A" for ref and alt because transcript position is not dependent on these values
# converts genomic position to transcript position
hgvs_name = str(pyhgvs.format_hgvs_name(chrom, genomicPos, "T", "A", genome, transcript))
# parses out transcript position from full hgvs_name
transcriptPos = str(pyhgvs.HGVSName(hgvs_name).cdna_start)
return transcriptPos
def test_apply_hgvs(self):
# p.Leu1303Phefs == c.3908dupT
transcripts = \
SeqIO.to_dict(SeqIO.parse("tests/test_data/BRCA1_transcripts.fa",
"fasta"))
brca1_mut = hgvs.HGVSName("ENST00000357654:c.3908dupT")
normal_p = transcripts["ENST00000357654"].seq.translate()
assert_equals("L", normal_p[1302])
mut_c = verify.apply_hgvs(transcripts["ENST00000357654"].seq,
brca1_mut)
assert_equals("TT", mut_c[3907:3909])
mut_p = mut_c.translate()
assert_equals("F", mut_p[1302])
def parseVar(variantHGVS):
'''
Parses the given variant HGVS and returns a dictionary containing:
HGVS type, variant type, ref allele, and alt allele
'''
varHGVS = hgvs.HGVSName(str(variantHGVS))
varParsed = {"typeHGVS": varHGVS.kind,
"varRef": varHGVS.ref_allele,
"varAlt": varHGVS.alt_allele,
"varType": varHGVS.mutation_type}
return varParsed
def get_hgvs_mutations(self, transcript_id, ignore_introns=True):
hgvs_muts = []
if ignore_introns:
df = self.df[self.df.Variant_Classification != "Intron"]
else:
df = self.df
for r in df.HGVS_coding_DNA_change:
if r != "Exception_encountered":
try:
h = hgvs.HGVSName(r)
# remove .version postfix
if h.transcript == transcript_id or ".".join(
h.transcript.split(".")[:-1]) == transcript_id:
hgvs_muts += [h]
except:
sys.stderr.write("Invalid HGVS found: {}\n".format(r))
pass
return hgvs_muts
def fix_mutalyzer(row, coord_hash):
if not pandas.isnull(row['Chromosomal Variant']):
return row['Chromosomal Variant'] + '___?'
# else mutalyzer cannot handle
try:
hgvs_name = hgvs.HGVSName(row['Input Variant'])
except:
return '___'
nm = hgvs_name.transcript
if not nm:
nm = row['Input Variant'].split(':')[0]
start = hgvs_name.cdna_start.coord
start_offset = hgvs_name.cdna_start.offset
end = hgvs_name.cdna_end.coord
end_offset = hgvs_name.cdna_end.offset
if row['Input Variant'] in ('NM_004985.4:c.*1638A>G',
'NM_001363.4:c.*6G>A',
'NM_004985.4:c.*2591A>G',
'NM_004985.4:c.*2888A>G',
'NM_004985.4:c.*3377C>T'):
return '___'
if start == end and start_offset == 0 and start > 0:
#print(nm, start, row['Input Variant'])
try:
chrom, g_coord, c_nuc, strand = coord_hash[nm][start]
if strand == '+':
ref, alt = hgvs_name.ref_allele, hgvs_name.alt_allele
else:
ref, alt = comp[hgvs_name.ref_allele], comp[
hgvs_name.alt_allele]
return 'XXX:g.%s%s>%s' % (g_coord, ref, alt) + '___' + chrom
except:
return '___'
return '___'
# 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:
#
# hgvs_name.transcript = 'NM_000352.3'
# hgvs_name.kind = 'c'
# hgvs_name.mutation_type = '>'
# hgvs_name.cdna_start = hgvs.CDNACoord(215, -10)
# hgvs_name.cdna_end = hgvs.CDNACoord(215, -10)
# hgvs_name.ref_allele = 'A'
# hgvs_name.alt_allele = 'G'
print((hgvs_name.transcript, hgvs_name.kind, hgvs_name.mutation_type,
hgvs_name.cdna_start, hgvs_name.cdna_end, hgvs_name.ref_allele,
hgvs_name.alt_allele))
def alter_coords_hgvs_sequential(h1, h2):
"""Change HGVS coords of h2 after applying h1"""
if h1.kind == "c" and h2.kind == "c":
if h1.mutation_type == ">":
h3 = hgvs.HGVSName(h2.name)
elif h1.mutation_type == "del":
if h1.cdna_start.coord > h2.cdna_end.coord:
h3 = hgvs.HGVSName(h2.name)
elif h1.cdna_end.coord < h2.cdna_start.coord:
h3 = hgvs.HGVSName(h2.name)
h3.cdna_start = hgvs.CDNACoord(coord=h3.cdna_start.coord -
len(h1.ref_allele))
h3.cdna_end = hgvs.CDNACoord(coord=h3.cdna_end.coord -
len(h1.ref_allele))
else:
raise (Exception(
"Overlapping del not implemented.\nhgvs1: {}\nhgvs2: {}".
format(h1, h2)))
elif h1.mutation_type == "ins":
if h1.cdna_start.coord > h2.cdna_end.coord:
h3 = hgvs.HGVSName(h2.name)
elif h1.cdna_end.coord < h2.cdna_start.coord:
h3 = hgvs.HGVSName(h2.name)
h3.cdna_start = hgvs.CDNACoord(coord=h3.cdna_start.coord +
len(h1.alt_allele))
h3.cdna_end = hgvs.CDNACoord(coord=h3.cdna_end.coord +
len(h1.alt_allele))
else:
raise (Exception("Overlapping ins not implemented"))
elif h1.mutation_type == "dup":
if h1.cdna_start.coord > h2.cdna_end.coord:
h3 = hgvs.HGVSName(h2.name)
elif h1.cdna_end.coord < h2.cdna_start.coord:
h3 = hgvs.HGVSName(h2.name)
h3.cdna_start = hgvs.CDNACoord(coord=h3.cdna_start.coord +
len(h1.alt_allele) -
len(h1.ref_allele))
h3.cdna_end = hgvs.CDNACoord(coord=h3.cdna_end.coord +
len(h1.alt_allele) -
len(h1.ref_allele))
else:
raise (Exception("Overlapping dup not implemented"))
elif h1.mutation_type == "delins":
if h1.cdna_start.coord > h2.cdna_end.coord:
h3 = hgvs.HGVSName(h2.name)
elif h1.cdna_end.coord < h2.cdna_start.coord:
h3 = hgvs.HGVSName(h2.name)
h3.cdna_start = hgvs.CDNACoord(coord=h3.cdna_start.coord -
len(h1.ref_allele) +
len(h1.alt_allele))
h3.cdna_end = hgvs.CDNACoord(coord=h3.cdna_end.coord -
len(h1.ref_allele) +
len(h1.alt_allele))
else:
raise (Exception("Overlapping delins not implemented"))
else:
raise (Exception("Unexpected mutation_type {}".format(
h1.mutation_type)))
return h3
else:
raise (Exception("Only cDNA mutations have been implemented"))
genome = SequenceFileDB(args.ref)
# Read RefSeq transcripts into a python dict.
#with open('/home/mbosio/projects/rtt/code/chrx.txt') as infile:
with open(args.transcript) 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) as rd,open(args.outfile,'w') as wr:
wr.write('\t'.join(['#Chr','Start','End','Ref','Alt','Group']) + '\n')
for line in rd:
#print line.strip()
try:
a = hgvs.HGVSName(line.strip())
a.chrom='X'
outlist = hgvs.get_vcf_allele(a,genome)
outlist = [str(x) for x in outlist]
outlist.append('Cases')
wr.write('\t'.join(outlist)+'\n')
except :
wr.write('Error:%s %s \n'%(line.strip(),sys.exc_info()[0]))
print line.strip()
print sys.exc_info()[0]
pass