def format_main(args):
# generate bedpe file
hout = open(args.output, 'w')
with open(args.result_file, 'r') as hin:
for line in hin:
if line.startswith("#"): continue
if utils.header_check(line.rstrip('\n')):
header_info.read(line.rstrip('\n'))
continue
F = line.rstrip('\n').split('\t')
if F[header_info.variant_type] in ["inversion", "translocation"]: continue
if abs(int(F[header_info.pos_1]) - int(F[header_info.pos_2])) > int(args.max_size_thres): continue
if F[header_info.variant_type] == "deletion":
ref_seq = my_seq.get_seq(args.reference, F[header_info.chr_1], int(F[header_info.pos_1]), int(F[header_info.pos_2]) - 1)
alt_seq = ref_seq[0] if F[header_info.inserted_seq] == "---" else ref_seq[0] + F[header_info.inserted_seq]
pos = F[header_info.pos_1]
elif F[header_info.variant_type] == "tandem_duplication":
alt_seq = my_seq.get_seq(args.reference, F[header_info.chr_1], int(F[header_info.pos_1]) - 1, int(F[header_info.pos_2]))
alt_seq = alt_seq if F[header_info.inserted_seq] == "---" else alt_seq + F[header_info.inserted_seq]
ref_seq = alt_seq[0]
pos = str(int(F[header_info.pos_1]) - 1)
print('\t'.join([F[header_info.chr_1], pos, '.', ref_seq, alt_seq, '.', "PASS", '.']), file = hout)
hout.close()
def nonB_DB_main(args):
from . import nonB_DB
all_nonB_DB_type = ["A_Phased_Repeat", "Direct_Repeat", "G_Quadruplex_Motif", "Inverted_Repeat",
"Mirror_Repeat", "Short_Tandem_Repeat", "Z_DNA_Motif"]
if not os.path.exists(args.result_file):
raise ValueError("file not exists: " + args.result_file)
nonB_DB_bed = args.nonB_DB
nonB_DB_tb = pysam.TabixFile(nonB_DB_bed)
hout = open(args.output, 'w')
with open(args.result_file, 'r') as hin:
for line in hin:
if utils.header_check(line.rstrip('\n')):
header_info.read(line.rstrip('\n'))
print_header = line.rstrip('\n') + '\t' + '\t'.join([x + "_dist1" + '\t' + x + "_dist2" for x in all_nonB_DB_type])
print(print_header, file = hout)
continue
F = line.rstrip('\n').split('\t')
# for meta info print
if F[0].startswith("#"):
print('\t'.join(F), file = hout)
continue
if utils.check_atypical_chromosomes(F[header_info.chr_1], F[header_info.chr_2]):
print("Skip a SV incolving atypical chromosomes: %s,%s,%s,%s,%s,%s" % \
(F[header_info.chr_1], F[header_info.pos_1], F[header_info.dir_1], \
F[header_info.chr_2], F[header_info.pos_2], F[header_info.dir_2]), file = sys.stderr)
continue
chr_ucsc1 = F[header_info.chr_1] if F[header_info.chr_1].startswith("chr") else "chr" + F[header_info.chr_1]
chr_ucsc2 = F[header_info.chr_2] if F[header_info.chr_2].startswith("chr") else "chr" + F[header_info.chr_2]
print_dist_bar = ''
for nonB_DB_type in all_nonB_DB_type:
nonB_DB_dist1 = nonB_DB.nonB_DB_dist_check(chr_ucsc1, int(F[header_info.pos_1]), F[header_info.dir_1], nonB_DB_tb, nonB_DB_type)
nonB_DB_dist2 = nonB_DB.nonB_DB_dist_check(chr_ucsc2, int(F[header_info.pos_2]), F[header_info.dir_2], nonB_DB_tb, nonB_DB_type)
print_dist_bar = print_dist_bar + '\t' + str(nonB_DB_dist1) + '\t' + str(nonB_DB_dist2)
print('\t'.join(F) + print_dist_bar, file = hout)
hout.close()
def AID_main(args):
# make directory for output if necessary
if os.path.dirname(args.output) != "" and not os.path.exists(os.path.dirname(args.output)):
os.makedirs(os.path.dirname(args.output))
hout = open(args.output, 'w')
with open(args.result_file, 'r') as hin:
for line in hin:
if line.startswith("#"): continue
if utils.header_check(line.rstrip('\n')):
line = line.rstrip('\n')
header_info.read(line)
print(line + '\t' + "CG_motif_info_1" + '\t' + "CG_motif_info_2" + '\t' + "WGCW_motif_info_1" + '\t' + "WGCW_motif_info_2", file = hout)
continue
F = line.rstrip('\n').split('\t')
if utils.check_atypical_chromosomes(F[header_info.chr_1], F[header_info.chr_2]):
print("Skip a SV incolving atypical chromosomes: %s,%s,%s,%s,%s,%s" % \
(F[header_info.chr_1], F[header_info.pos_1], F[header_info.dir_1], \
F[header_info.chr_2], F[header_info.pos_2], F[header_info.dir_2]), file = sys.stderr)
continue
seq1 = my_seq.get_seq(args.reference, F[header_info.chr_1], int(F[header_info.pos_1]) - args.check_size, int(F[header_info.pos_1]) + args.check_size)
seq2 = my_seq.get_seq(args.reference, F[header_info.chr_2], int(F[header_info.pos_2]) - args.check_size, int(F[header_info.pos_2]) + args.check_size)
CG_starts_1 = [match.start() - 10 for match in re.finditer(r'CG', seq1)]
CG_starts_2 = [match.start() - 10 for match in re.finditer(r'CG', seq2)]
WGCW_starts_1 = [match.start() - 10 for match in re.finditer(r'[AT]GC[AT]', seq1)]
WGCW_starts_2 = [match.start() - 10 for match in re.finditer(r'[AT]GC[AT]', seq2)]
if len(CG_starts_1) == 0: CG_starts_1.append("---")
if len(CG_starts_2) == 0: CG_starts_2.append("---")
if len(WGCW_starts_1) == 0: WGCW_starts_1.append("---")
if len(WGCW_starts_2) == 0: WGCW_starts_2.append("---")
print('\t'.join(F) + '\t' + \
','.join([str(x) for x in CG_starts_1]) + '\t' + \
','.join([str(x) for x in CG_starts_2]) + '\t' + \
','.join([str(x) for x in WGCW_starts_1]) + '\t' + \
','.join([str(x) for x in WGCW_starts_2]), file = hout)
hout.close()
def homology_main(args):
from . import homology
hout = open(args.output, 'w')
with open(args.result_file, 'r') as hin:
for line in hin:
if line.startswith("#"): continue
if utils.header_check(line.rstrip('\n')):
header_info.read(line.rstrip('\n'))
print_header = line.rstrip('\n') + '\t' + "Homology_Match"
print(print_header, file = hout)
continue
F = line.rstrip('\n').split('\t')
# for meta info print
if F[0].startswith("#"):
print('\t'.join(F), file = hout)
continue
if utils.check_atypical_chromosomes(F[header_info.chr_1], F[header_info.chr_2]):
print("Skip a SV incolving atypical chromosomes: %s,%s,%s,%s,%s,%s" % \
(F[header_info.chr_1], F[header_info.pos_1], F[header_info.dir_1], \
F[header_info.chr_2], F[header_info.pos_2], F[header_info.dir_2]), file = sys.stderr)
continue
var_size = 500000
if F[header_info.variant_type] == "deletion":
var_size = int(F[header_info.pos_2]) - int(F[header_info.pos_1]) - 1
elif F[header_info.variant_type] == "tandem_duplication":
var_size = int(F[header_info.pos_2]) - int(F[header_info.pos_1]) + 1
homology_match = homology.check_homology(F[header_info.chr_1], F[header_info.pos_1], F[header_info.dir_1],
F[header_info.chr_2], F[header_info.pos_2], F[header_info.dir_2],
args.reference, min(var_size, 100))
print('\t'.join(F) + '\t' + str(homology_match), file = hout)
hout.close()
def RSS_main(args):
# make directory for output if necessary
if os.path.dirname(args.output) != "" and not os.path.exists(os.path.dirname(args.output)):
os.makedirs(os.path.dirname(args.output))
rss_pwm = my_seq.generate_rss_pwm()
hout = open(args.output, 'w')
with open(args.result_file, 'r') as hin:
for line in hin:
if line.startswith("#"): continue
if utils.header_check(line.rstrip('\n')):
line = line.rstrip('\n')
header_info.read(line)
print(line + '\t' + "RSS_score_1" + '\t' + "RSS_info_1" + '\t' + "RSS_score_2" + '\t' + "RSS_info_2", file = hout)
continue
F = line.rstrip('\n').split('\t')
if utils.check_atypical_chromosomes(F[header_info.chr_1], F[header_info.chr_2]):
print("Skip a SV incolving atypical chromosomes: %s,%s,%s,%s,%s,%s" % \
(F[header_info.chr_1], F[header_info.pos_1], F[header_info.dir_1], \
F[header_info.chr_2], F[header_info.pos_2], F[header_info.dir_2]), file = sys.stderr)
continue
seq1 = my_seq.get_seq(args.reference, F[header_info.chr_1], int(F[header_info.pos_1]) - args.check_size, int(F[header_info.pos_1]) + args.check_size)
seq2 = my_seq.get_seq(args.reference, F[header_info.chr_2], int(F[header_info.pos_2]) - args.check_size, int(F[header_info.pos_2]) + args.check_size)
rss_info_1 = my_seq.get_max_rss_score(seq1, rss_pwm[0], rss_pwm[1])
rss_info_2 = my_seq.get_max_rss_score(seq2, rss_pwm[0], rss_pwm[1])
print('\t'.join(F) + '\t' + \
str(round(rss_info_1[0], 3)) + '\t' + \
';'.join([rss_info_1[1], rss_info_1[2], str(int(rss_info_1[3] - 50)), str(rss_info_1[4]), str(rss_info_1[5])]) + '\t' + \
str(round(rss_info_2[0], 3)) + '\t' + \
';'.join([rss_info_2[1], rss_info_2[2], str(int(rss_info_2[3] - 50)), str(rss_info_2[4]), str(rss_info_2[5])]), file = hout)
hout.close()
def primer_main(args):
from genomon_sv import realignmentFunction
from primer3 import bindings
# make directory for output if necessary
if os.path.dirname(args.output) != "" and not os.path.exists(os.path.dirname(args.output)):
os.makedirs(os.path.dirname(args.output))
param = {"reference_genome": args.reference, "split_refernece_thres": 1000, "validate_sequence_length": 250}
hout = open(args.output, 'w')
with open(args.result_file, 'r') as hin:
for line in hin:
if line.startswith("#"): continue
if utils.header_check(line.rstrip('\n')):
line = line.rstrip('\n')
header_info.read(line)
print(line + '\t' + "Primer1" + '\t' + "Primer2" + '\t' + "Primer3" + '\t' + "Primer4" + '\t' + "Primer5", file = hout)
continue
F = line.rstrip('\n').split('\t')
chr1, pos1, dir1, chr2, pos2, dir2, junc_seq = F[header_info.chr_1], F[header_info.pos_1], F[header_info.dir_1], \
F[header_info.chr_2], F[header_info.pos_2], F[header_info.dir_2], F[header_info.inserted_seq]
if utils.check_atypical_chromosomes(chr1, chr2):
print("Skip a SV incolving atypical chromosomes: %s,%s,%s,%s,%s,%s" % \
(chr1, pos1, dir1, chr2, pos2, dir2), file = sys.stderr)
continue
junc_seq_len = 0 if junc_seq == "---" else len(junc_seq)
realignmentFunction.getRefAltForSV(args.output + ".contig.tmp.fa", chr1, pos1, dir1, chr2, pos2, dir2, junc_seq, args.reference, 1000, 250)
with open(args.output + ".contig.tmp.fa") as hin2:
lines2 = hin2.readlines()
for i in range(len(lines2)):
lines2[i] = lines2[i].rstrip('\n')
if lines2[i].startswith('>') and lines2[i].endswith("alt"):
seq = lines2[i + 1].rstrip('\n')
primer = bindings.designPrimers(
{
'SEQUENCE_ID': 'MH1000',
'SEQUENCE_TEMPLATE': seq,
'SEQUENCE_TARGET': [225,50 + junc_seq_len],
'SEQUENCE_INCLUDED_REGION': [10, len(seq) - 20]
},
{
'PRIMER_PRODUCT_SIZE_RANGE': [[150,250],[100,300],[301,400],[401,500]],
})
primer_left_right = ["---"] * 5
for i in range(5):
if "PRIMER_LEFT_" + str(i) + "_SEQUENCE" in primer and "PRIMER_RIGHT_" + str(i) + "_SEQUENCE" in primer and \
"PRIMER_LEFT_" + str(i) + "_TM" in primer and "PRIMER_RIGHT_" + str(i) + "_TM" in primer and \
"PRIMER_PAIR_" + str(i) + "_PRODUCT_SIZE" in primer:
primer_left_right[i] = primer["PRIMER_LEFT_" + str(i) + "_SEQUENCE"] + ";" + primer["PRIMER_RIGHT_" + str(i) + "_SEQUENCE"] + ';' + \
str(round(primer["PRIMER_LEFT_" + str(i) + "_TM"], 3)) + ";" + str(round(primer["PRIMER_RIGHT_" + str(i) + "_TM"], 3)) + ';' + \
str(primer["PRIMER_PAIR_" + str(i) + "_PRODUCT_SIZE"])
print('\t'.join(F) + '\t' + '\t'.join(primer_left_right), file = hout)
hout.close()
subprocess.check_call(["rm", "-rf", args.output + ".contig.tmp.fa"])
def realign_main(args):
from genomon_sv import filterFunction
if args.tumor_bam is None:
print("tumor_bam file should be input", file = sys.stderr)
sys.exit(1)
# make directory for output if necessary
if os.path.dirname(args.output) != "" and not os.path.exists(os.path.dirname(args.output)):
os.makedirs(os.path.dirname(args.output))
matchedControlFlag = True if args.control_bam is not None else False
if args.control_bam is None: args.control_bam = ""
# generate bedpe file
hout = open(args.output + ".tmp1.bedpe", 'w')
i = 0
with open(args.result_file, 'r') as hin:
for line in hin:
if line.startswith("#"): continue
if utils.header_check(line.rstrip('\n')):
line = line.rstrip('\n')
header_info.read(line)
continue
F = line.rstrip('\n').split('\t')
if utils.check_atypical_chromosomes(F[header_info.chr_1], F[header_info.chr_2]):
print("Skip a SV incolving atypical chromosomes: %s,%s,%s,%s,%s,%s" % \
(F[header_info.chr_1], F[header_info.pos_1], F[header_info.dir_1], \
F[header_info.chr_2], F[header_info.pos_2], F[header_info.dir_2]), file = sys.stderr)
continue
print('\t'.join([F[header_info.chr_1], str(int(F[header_info.pos_1]) - 1), F[header_info.pos_1], \
F[header_info.chr_2], str(int(F[header_info.pos_2]) - 1), F[header_info.pos_2], \
"genoemonSV_" + str(i), F[header_info.inserted_seq], F[header_info.dir_1], F[header_info.dir_2]] + \
["---" for i in range(14)]), file = hout)
i = i + 1
hout.close()
filterFunction.validateByRealignment(args.output + ".tmp1.bedpe",
args.output + ".tmp2.bedpe",
args.tumor_bam,
args.control_bam,
args.reference,
"-stepSize=5 -repMatch=2253",
500,
5000,
1000,
5,
1000,
1000)
key2AF_info = {}
with open(args.output + ".tmp2.bedpe", 'r') as hin:
for line in hin:
F = line.rstrip('\n').split('\t')
key = '\t'.join(F[:7])
tumorAF = 0.0
if float(F[7]) + float(F[8]) > 0: tumorAF = float(F[8]) / (float(F[7]) + float(F[8]))
tumorAF = str(round(tumorAF, 4))
normalAF = "---"
if matchedControlFlag == True:
normalAF = 0.0
if float(F[9]) + float(F[10]) > 0: normalAF = float(F[10]) / (float(F[9]) + float(F[10]))
normalAF = str(round(normalAF, 4))
if matchedControlFlag == True:
key2AF_info[key] = '\t'.join([F[7], F[8], tumorAF, F[9], F[10], normalAF, F[11]])
else:
key2AF_info[key] = '\t'.join([F[7], F[8], tumorAF])
hout = open(args.output, 'w')
with open(args.result_file, 'r') as hin:
for line in hin:
if line.startswith("#"): continue
if utils.header_check(line.rstrip('\n')):
line = line.rstrip('\n')
if matchedControlFlag == True:
print(line + '\t' + "Num_Tumor_Ref_Read_Pair_re" + '\t' + "Num_Tumor_Var_Read_Pair_re" + '\t' + "Tumor_VAF_re" + '\t' + \
"Num_Control_Ref_Read_Pair_re" + '\t'+ "Num_Control_Var_Read_Pair_re" + '\t' + "Control_VAF_re" + '\t' + \
"Minus_Log_Fisher_P_value_re", file = hout)
else:
print(line + '\t' + "Num_Tumor_Ref_Read_Pair_re" + '\t' + "Num_Tumor_Var_Read_Pair_re" + '\t' + "Tumor_VAF_re", file = hout)
continue
F = line.rstrip('\n').split('\t')
key = '\t'.join(F[:7])
if key not in key2AF_info: continue
print('\t'.join(F) + '\t' + key2AF_info[key], file = hout)
hout.close()
subprocess.check_call(["rm", "-rf", args.output + ".tmp1.bedpe"])
subprocess.check_call(["rm", "-rf", args.output + ".tmp2.bedpe"])
def merge_control_main(args):
import genomon_sv.mergeFunction, genomon_sv.utils
# make directory for output if necessary
if os.path.dirname(args.output_file) != "" and not os.path.exists(os.path.dirname(args.output_file)):
os.makedirs(os.path.dirname(args.output_file))
hout = open(args.output_file + ".temp", 'w')
tumor_type_list = {}
gene2type_sample = {}
with open(args.result_list, 'r') as hin:
for line in hin:
label, tumor_type, result_file = line.rstrip('\n').split('\t')
# label, result_file = line.rstrip('\n').split('\t')
if tumor_type not in tumor_type_list: tumor_type_list[tumor_type] = 1
if not os.path.exists(result_file):
raise ValueError("file not exists: " + result_file)
num = 1
with open(result_file, 'r') as hin:
for line in hin:
if line.startswith("#"): continue
if utils.header_check(line.rstrip('\n')):
line = line.rstrip('\n')
header_info.read(line)
continue
F = line.rstrip('\n').split('\t')
inseqLen = len(F[header_info.inserted_seq]) if F[header_info.inserted_seq] != "---" else 0
print('\t'.join([F[header_info.chr_1], str(int(F[header_info.pos_1]) - 1), F[header_info.pos_1], \
F[header_info.chr_2], str(int(F[header_info.pos_2]) - 1), F[header_info.pos_2], \
"junction_" + str(num), str(inseqLen), \
F[header_info.dir_1], F[header_info.dir_2], label, "1"]), file = hout)
num = num + 1
hout.close()
# utils.processingMessage("sorting the aggregated junction file")
genomon_sv.utils.sortBedpe(args.output_file + ".temp", args.output_file + ".temp.sort")
# utils.processingMessage("merging the same junction in the aggregated junction file")
genomon_sv.mergeFunction.organizeControl(args.output_file + ".temp.sort", args.output_file + ".temp.merged", 20)
# utils.processingMessage("sorting the merged junction file")
genomon_sv.utils.sortBedpe(args.output_file + ".temp.merged", args.output_file + ".temp.merged.sort")
# utils.processingMessage("compressing the merged junction file")
genomon_sv.utils.compress_index_bed(args.output_file + ".temp.merged.sort", args.output_file)
# remove intermediate files
subprocess.check_call(["rm", "-rf", args.output_file + ".temp"])
subprocess.check_call(["rm", "-rf", args.output_file + ".temp.sort"])
subprocess.check_call(["rm", "-rf", args.output_file + ".temp.merged"])
subprocess.check_call(["rm", "-rf", args.output_file + ".temp.merged.sort"])