def test_mutate_indel():
from Bio.Seq import Seq
from Bio.SeqRecord import SeqRecord
from Bio.Alphabet import generic_dna
from common_object import Boundary, Variant
seq = SeqRecord(Seq('CCTGGTGCTC', generic_dna))
reg = Boundary('chr1', 0, 10)
var = Variant('chr1', 2, 3, 'indel', 'ins', 'T', 'TAC')
reg.variants.append(var)
trueSeq = mutate_indel(reg, var, seq)
assert (str(trueSeq) == 'CCTACGGTGCTC')
'''-------------'''
seq = SeqRecord(Seq('CCTGGTGCTC', generic_dna))
reg = Boundary('chr1', 0, 10)
var = Variant('chr1', 12, 13, 'indel', 'ins', 'T', 'TAC')
reg.variants.append(var)
trueSeq = mutate_indel(reg, var, seq)
assert (str(trueSeq) == 'CCTGGTGCTC')
'''---------------'''
seq = SeqRecord(Seq('CCTGGTGCTC', generic_dna))
reg = Boundary('chr1', 0, 10)
var = Variant('chr1', 2, 2, 'indel', 'ins', 'T', ['TAC', 'A'])
reg.variants.append(var)
trueSeq = mutate_indel(reg, var, seq)
assert (str(trueSeq) == 'CCTACGGTGCTC')
'''----------------'''
seq = SeqRecord(Seq('CCTGGTGCTC', generic_dna))
reg = Boundary('chr1', 0, 10)
var = Variant('chr1', 2, 3, 'indel', 'del', 'TG', 'T')
reg.variants.append(var)
trueSeq = mutate_indel(reg, var, seq)
assert (str(trueSeq) == 'CCTGTGCTC')
def test_mutate_sv():
from Bio.Seq import Seq
from Bio.SeqRecord import SeqRecord
from Bio.Alphabet import generic_dna
from common_object import Boundary, Variant
seq = SeqRecord(Seq('CCTGGTGCTC', generic_dna))
reg = Boundary('chr1', 0, 10)
var = Variant('chr1', 1, 5, 'sv', 'DEL', '', '')
# seq = CCTGGTGCTC, NNCTGCTCNN
reg.variants.append(var)
trueSeq = mutate_sv(reg, var, seq)
assert (str(trueSeq) == 'CTGCTC')
'''-----------------------------'''
seq = SeqRecord(Seq('CCTGGTGCTC', generic_dna))
reg = Boundary('chr1', 0, 10)
var = Variant('chr1', 0, 6, 'sv', 'INV', '', '')
reg.variants.append(var)
trueSeq = mutate_sv(reg, var, seq)
assert (str(trueSeq) == 'TGGTCCGCTC')
'''-----------------------------'''
seq = SeqRecord(Seq('CCTGGTGCTC', generic_dna))
reg = Boundary('chr1', 0, 10)
var = Variant('chr1', 0, 3, 'sv', 'DUP', '', '')
reg.variants.append(var)
trueSeq = mutate_sv(reg, var, seq)
assert (str(trueSeq) == 'CCTCCTGGTGCTC')
'''-----------------------------'''
seq = SeqRecord(Seq('CCTGGTGCTC', generic_dna))
reg = Boundary('chr1', 0, 10)
var = Variant('chr1', 0, 3, 'sv', 'CNV', '', alt='[3,3]')
reg.variants.append(var)
trueSeq = mutate_sv(reg, var, seq)
assert (str(trueSeq) == 'CCTCCTCCTGGTGCTC')
def get_ssm_bedfile(ssm_file: str, cohort: str = 'Eric_CLL'):
"""Obtains and parses the simple somatic mutation file bed file
Parameters
----------
ssm_file : the simple somatic mutation file in a bed file format
cohort : a field in the field labeled as cohort
Returns
-------
variants: a dictionary of list of Variant objects
"""
cohort = cohort.replace('Eric_', '')
variants = {}
count = 0
with open(ssm_file) as fi:
for ln in fi.readlines():
st = re.split('[\t\n]+', ln)
# print('line:', ln)
# print('st:', st)
if st[6] != cohort:
continue
chrom = st[0]
start = int(st[1])
end = int(st[2])
sample_id = st[3]
if sample_id not in variants:
variants[sample_id] = []
ref = st[4]
alt = st[5]
vt = 'snp'
svtype = None
gt = '1|1'
count += 1
var = Variant(chrom, start, end, vt, svtype, ref, alt, gt)
variants[sample_id].append(var)
print('number of samples:{}, number of mutation:{}, mutations/sample:{}'.
format(len(variants), count,
float(count) / len(variants)))
return variants
def test_mutate_snp():
from Bio.Seq import Seq
from Bio.SeqRecord import SeqRecord
from Bio.Alphabet import generic_dna
from common_object import Boundary, Variant
seq = SeqRecord(Seq('CCTGGTGCTC', generic_dna))
reg = Boundary('chr1', 0, 10)
var = Variant('chr1', 2, 2, 'snp', '', 'T', 'A')
reg.variants.append(var)
true_seq = mutate_snp(reg, var, seq)
assert str(true_seq) == 'CCAGGTGCTC'
reg = Boundary('chr1', 0, 10)
var = Variant('chr1', 1, 1, 'snp', '', 'C', 'AT')
reg.variants.append(var)
true_seq = mutate_snp(reg, var, seq)
assert str(true_seq) == 'CATTGGTGCTC'
def getSSM_bedfile(ssm_file, cohort='Eric_CLL'):
cohort = cohort.replace('Eric_','')
variants = {}
count = 0
with open(ssm_file) as fi:
for ln in fi.readlines():
st = re.split('[\t\n]+',ln)
# print('line:', ln)
# print('st:', st)
if st[6] != cohort:
continue
chrom = st[0]
start = int(st[1])
end = int(st[2])
sample_id = st[3]
if sample_id not in variants:
variants[sample_id] = []
ref = st[4]
alt = st[5]
vt = 'snp'
svtype = None
gt = '1|1'
count += 1
var = Variant(chrom, start, end, vt, svtype, ref, alt, gt)
variants[sample_id].append(var)
print('number of samples:{}, number of mutation:{}, mutations/sample:{}'.format(len(variants), count,
float(count)/len(variants)))
return variants
def getSV(stvm_file):
variants = {}
processed_sv = set()
with open(stvm_file, 'r') as fin:
ln = fin.readline()
fields = re.split('\t', ln)
field2Id = {}
for i in range(len(fields)):
field2Id[fields[i]] = i
sv_header_id = field2Id['sv_id']
variant_type_id = field2Id['variant_type']
chr_from_id = field2Id['chr_from']
chr_to_id = field2Id['chr_to']
icgc_sample_id = field2Id['icgc_sample_id']
chr_from_bkpt_id = field2Id['chr_from_bkpt']
chr_to_bkpt_id = field2Id['chr_to_bkpt']
for ln in fin.readlines():
st = re.split('\t', ln)
sv_id = st[sv_header_id]
if sv_id in processed_sv:
continue
processed_sv.add(sv_id)
svtype = st[variant_type_id]
# if svtype == 'unbalanced translocation':
# continue
chrid_from = st[chr_from_id].upper()
chrid_to = st[chr_to_id].upper()
# ignore inter-chromosome SV
if (chrid_from != chrid_to) or (not re.search('^[0-9XY]+', chrid_from)) \
or (not re.search('^[0-9XY]+', chrid_to)):
continue
sample_id = st[icgc_sample_id]
if sample_id not in variants:
variants[sample_id] = []
chrom = 'chr' + chrid_from
start = int(st[chr_from_bkpt_id]) - 1
end = int(st[chr_to_bkpt_id])
if svtype == 'deletion':
svtype = 'DEL'
elif svtype == 'inversion':
svtype = 'INV'
elif svtype == 'tandem duplication':
svtype = 'DUP'
else:
sys.stderr.write('wrong svtype:{}\n'.format(svtype))
ref = ''
alt = ''
gt = '1|1'
vt = 'sv'
var = Variant(chrom, start, end, vt, svtype, ref, alt, gt)
variants[sample_id].append(var)
for sample, vars in variants.items():
for i in range(len(vars) - 1):
if vars[i].chrid == vars[i + 1].chrid and vars[i].start == vars[i + 1].start\
and vars[i].end == vars[i + 1].end:
print('duplicate variant, sample:{}, variants:{}'.format(sample, str(vars[i])))
print('Number of SV samples:', len(variants))
return variants
def getSSM(ssm_file):
variants = {} # icgc_sample_id: variants
processed_mut = set() # processsed mutation to handle duplicate records
# for i in range(len(ssm)):
with open(ssm_file, 'r') as fin:
ln = fin.readline()
fields = re.split('\t', ln)
field2Id = {}
for i in range(len(fields)):
field2Id[fields[i]] = i
icgc_mutation_id = field2Id['icgc_mutation_id']
chromosome_id = field2Id['chromosome']
chromosome_start_id = field2Id['chromosome_start']
chromosome_end_id = field2Id['chromosome_end']
mutation_type_id = field2Id['mutation_type']
reference_genome_allele_id = field2Id['reference_genome_allele']
if 'tumour_genotype' in fields:
tumour_genotype_id = field2Id['tumour_genotype']
else:
mutated_to_allele_id = field2Id['mutated_to_allele']
icgc_sample_id = field2Id['icgc_sample_id']
for ln in fin.readlines():
st = re.split('\t', ln)
mut_id = st[icgc_mutation_id]
if mut_id in processed_mut:
#print(mut_id)
continue
processed_mut.add(mut_id)
chrid = str(st[chromosome_id]).upper()
if not re.search('^[0-9XY]+', chrid):
print(chrid)
continue
start = int(st[chromosome_start_id]) - 1
end = int(st[chromosome_end_id])
vt = st[mutation_type_id]
ref = st[reference_genome_allele_id]
if 'tumour_genotype' in fields:
# control_gt = ssm.loc[i, 'control_genotype']
tumor_gt = st[tumour_genotype_id]
else:
tumor_gt = st[mutated_to_allele_id]
sample_id = st[icgc_sample_id]
if sample_id not in variants:
variants[sample_id] = []
'''
mutating by replacing ref. with alt.
if alt == '', in insertation, it means no insertation ( if ref == '')
in deletion, it means deleting ref
if insertion, ref is always -, must be converted to ''
'''
# Variant(sample, rc.CHROM, start, end, dvt, dsvtype, rc.REF, rc.ALT, gt )
ref = '' if ref == '-' else ref
if re.search('substitution', vt):
vt = 'snp'
svtype = ''
elif re.search('deletion', vt):
vt = 'indel'
svtype = 'del'
elif re.search('insertion', vt):
vt = 'indel'
svtype = 'ins'
# alternative
alt = re.split('[|/]', tumor_gt)
alt = [x if x != '-' else '' for x in alt] # convert '-' to empty
if len(alt) == 1: # if there is only one allele, make another from it
alt.append(alt[0])
# 0: for reference seq, therefore + 1
# if insertion, gt = '' can be 0
gt = '|'.join([str(alt.index(x) + 1) for x in alt])
if vt == 'snp' and ref == '':
print('error, snp but ref. is not available')
#print('chrom:{}, start:{}, end:{}, vt:{}, subtype:{}, ref:{}, alt:{}'.format(chrid, start, end, vt, svtype, ref, alt))
var = Variant('chr' + chrid, start, end, vt, svtype, ref, alt, gt)
variants[sample_id].append(var)
print('Number of sample:{}, number of variants{}'.format(len(variants), len(processed_mut)))
return variants
def get_sv(stvm_file: str) -> Dict:
"""Obtains and parses the structural variant file from ICGC
Parameters
----------
stvm_file : the structural variant file from ICGC in a tsv file format
Returns
-------
variants: a dictionary of list of Variant objects
"""
variants = {}
processed_sv = set()
with open(stvm_file, 'r') as fin:
ln = fin.readline()
fields = re.split('\t', ln)
field2_id = {}
# for i in range(len(fields)):
# field2Id[fields[i]] = i
# enumerate approach
for i, field in enumerate(fields):
field2_id[field] = i
sv_header_id = field2_id['sv_id']
variant_type_id = field2_id['variant_type']
chr_from_id = field2_id['chr_from']
chr_to_id = field2_id['chr_to']
icgc_sample_id = field2_id['icgc_sample_id']
chr_from_bkpt_id = field2_id['chr_from_bkpt']
chr_to_bkpt_id = field2_id['chr_to_bkpt']
for ln in fin.readlines():
st = re.split('\t', ln)
sv_id = st[sv_header_id]
if sv_id in processed_sv:
continue
processed_sv.add(sv_id)
svtype = st[variant_type_id]
# if svtype == 'unbalanced translocation':
# continue
chrid_from = st[chr_from_id].upper()
chrid_to = st[chr_to_id].upper()
# ignore inter-chromosome SV
if (chrid_from != chrid_to) or (
not re.search('^[0-9XY]+', chrid_from)) or \
(not re.search('^[0-9XY]+', chrid_to)):
continue
sample_id = st[icgc_sample_id]
if sample_id not in variants:
variants[sample_id] = []
chrom = 'chr' + chrid_from
start = int(st[chr_from_bkpt_id]) - 1
end = int(st[chr_to_bkpt_id])
if svtype == 'deletion':
svtype = 'DEL'
elif svtype == 'inversion':
svtype = 'INV'
elif svtype == 'tandem duplication':
svtype = 'DUP'
else:
sys.stderr.write('wrong svtype:{}\n'.format(svtype))
ref = ''
alt = ''
gt = '1|1'
vt = 'sv'
var = Variant(chrom, start, end, vt, svtype, ref, alt, gt)
variants[sample_id].append(var)
for sample, varts in variants.items():
for i in range(len(varts) - 1):
if varts[i].chrid == varts[i + 1].chrid and \
varts[i].start == varts[i + 1].start and \
varts[i].end == varts[i + 1].end:
print('duplicate variant, sample:{}, variants:{}'.format(
sample, str(varts[i])))
print('Number of SV samples:', len(variants))
return variants