def knownRegions(refdict, chromosomes):
ends = {c: None for c in chromosomes}
assert os.path.isfile(refdict)
with open(refdict, 'r') as file:
for line in file:
if "@SQ" in line:
assert "SN:" in line and "LN:" in line
c = line.split('SN:')[1].split()[0]
if c[3:] != "chr":
c = "chr{}".format(c)
if c in chromosomes:
end = int(line.split('LN:')[1].split()[0])
ends[c] = end
'''
with open(refdict, 'r') as i:
for l in i:
if '@SQ' in l:
assert 'SN:' in l and 'LN:' in l
c = l.split('SN:')[1].split()[0]
if c in chromosomes:
end = int(l.split('LN:')[1].split()[0])
ends[c] = end
'''
if None in ends.values():
log(msg=
"The following chromosomes have not been found in the dictionary of the reference genome: \n\t{}"
.format(','.join([c for c in ends if ends[c] is None])),
level="WARN")
res = {}
for c in chromosomes:
res[c] = [(0, ends[c])]
return res
def main():
log(msg=
"# Parsing the input arguments, checking the consistency of given files, and extracting required "
"information\n",
level="STEP")
args = ap.parse_baf_arguments()
logArgs(args, 80)
if args["reference"]:
defaultMode(args)
else:
naiveMode(args)
def defaultMode(args):
log(msg="# Inferring SNPs from the normal sample\n", level="STEP")
snps = SNPCalling.call(samtools=args["samtools"],
bcftools=args["bcftools"],
reference=args["reference"],
samples=[args["normal"]],
chromosomes=args["chromosomes"],
num_workers=args["j"],
snplist=args["snps"],
q=args["q"],
Q=args["Q"],
qual=args["qual"],
mincov=args["mincov"],
dp=args["maxcov"],
E=args["E"],
regions=args["regions"],
verbose=args["verbose"])
if not snps:
sp.close("No SNPs found in the normal!\n")
log(msg="# Selecting heterozygous SNPs\n", level="STEP")
hetSNPs = selectHetSNPs(counts=snps,
gamma=args["gamma"],
maxshift=args["maxshift"],
verbose=args["verbose"])
if not hetSNPs:
sp.close(
"No heterozygous SNPs found in the selected regions of the normal!\n"
)
log(msg=
"# Writing the list of selected SNPs, covered and heterozygous in the normal sample\n",
level="STEP")
with open(args["outputSnps"], 'w') as f:
for chro in args["chromosomes"]:
if (args["normal"][1], chro) in hetSNPs:
for snp in hetSNPs[args["normal"][1], chro]:
f.write("{}\t{}\n".format(snp[1], snp[2]))
log(msg=
"# Writing the allele counts of the normal sample for selected SNPs\n",
level="STEP")
if args["outputNormal"] is not None:
with open(args["outputNormal"], 'w') as f:
for chro in args["chromosomes"]:
if (args["normal"][1], chro) in hetSNPs:
for count in hetSNPs[args["normal"][1], chro]:
f.write("{}\t{}\t{}\t{}\t{}\n".format(
count[0], count[1], count[2], count[3], count[4]))
else:
for chro in args["chromosomes"]:
if (args["normal"][1], chro) in hetSNPs:
for count in hetSNPs[args["normal"][1], chro]:
sys.stdout.write("{}\t{}\t{}\t{}\t{}\n".format(
count[0], count[1], count[2], count[3], count[4]))
log(msg="# Counting the alleles of tumor samples for selected SNPs\n",
level="STEP")
counts = \
AlleleCounting.count(samtools=args["samtools"], bcftools=args["bcftools"], reference=args["reference"],
samples=args["samples"], chromosomes=args["chromosomes"], num_workers=args["j"],
snplist=args["outputSnps"], q=args["q"], Q=args["Q"], mincov=args["mincov"],
dp=args["maxcov"], E=args["E"], verbose=args["verbose"])
if not counts:
sp.close("The selected SNPs are not covered in the tumors!\n")
log(msg="# Writing the allele counts of tumor samples for selected SNPs\n",
level="STEP")
if args["outputTumors"] is not None:
with open(args["outputTumors"], 'w') as f:
for sample in args["samples"]:
for chro in args["chromosomes"]:
if (sample[1], chro) in counts:
for count in counts[sample[1], chro]:
f.write("{}\t{}\t{}\t{}\t{}\n".format(
count[0], count[1], count[2], count[3],
count[4]))
else:
for sample in args["samples"]:
for chro in args["chromosomes"]:
if (sample[1], chro) in counts:
for count in counts[sample[1], chro]:
sys.stdout.write("{}\t{}\t{}\t{}\t{}\n".format(
count[0], count[1], count[2], count[3], count[4]))
def logArgs(args, width):
text = "\n"
for key in args:
text += "\t{}: {}\n".format(key, args[key])
log(msg=text, level="INFO")
def main():
log(msg="# Parsing and checking input arguments\n", level="STEP")
args = ap.parse_bin_arguments()
logArgs(args, 80)
if args["regions"] is None:
log(msg=
"# Retrieving genomic regions to consider from maximum chromosome length\n",
level="STEP")
regions = knownRegions(args["refdict"], args["chromosomes"])
else:
log(msg="# Checking the consistency of the given regions\n",
level="STEP")
regions = ap.parseRegions(args["regions"], args["chromosomes"])
if args["verbose"]:
msg = "regions: "
for c in args["chromosomes"]:
msg += " {}: {}".format(c, regions[c])
msg += "\n"
log(msg=msg, level="INFO")
log(msg="# Binning and counting the normal sample\n", level="STEP")
normal_bins = \
bb.bin(samtools=args["samtools"], samples=[args["normal"]], chromosomes=args["chromosomes"],
num_workers=args["j"], q=args["q"], size=args["size"], regions=regions, verbose=args["verbose"])
if not normal_bins:
raise SystemExit("No bins in the normal sample!\n")
log(msg="# Writing the read counts for bins of normal sample\n",
level="STEP")
if args["outputNormal"] is not None:
with open(args["outputNormal"], 'w') as f:
for c in args["chromosomes"]:
for count in normal_bins[args["normal"][1], c]:
f.write("{}\t{}\t{}\t{}\t{}\n".format(
count[0], count[1], count[2], count[3], count[4]))
else:
for c in args["chromosomes"]:
for count in normal_bins[args["normal"][1], c]:
sys.stdout.write("{}\t{}\t{}\t{}\t{}\n".format(
count[0], count[1], count[2], count[3], count[4]))
log(msg="# Binning and counting the tumor samples\n", level="STEP")
tumor_bins = \
bb.bin(samtools=args["samtools"], samples=args["samples"], chromosomes=args["chromosomes"],
num_workers=args["j"], q=args["q"], size=args["size"], regions=regions, verbose=args["verbose"])
if not tumor_bins:
log.close("No bins in the tumor samples!\n")
log(msg="# Writing the read counts for bins of tumor samples\n",
level="STEP")
if args["outputTumors"] is not None:
with open(args["outputTumors"], 'w') as f:
for sample in args["samples"]:
for c in args["chromosomes"]:
for count in tumor_bins[sample[1], c]:
f.write("{}\t{}\t{}\t{}\t{}\n".format(
count[0], count[1], count[2], count[3], count[4]))
else:
for sample in args["samples"]:
for c in args["chromosomes"]:
for count in tumor_bins[sample[1], c]:
sys.stdout.write("{}\t{}\t{}\t{}\t{}\n".format(
count[0], count[1], count[2], count[3], count[4]))
log(msg="# Counting total number of reads for normal and tumor samples\n",
level="STEP")
total_counts = \
tc.tcount(samtools=args["samtools"], samples=({args["normal"]}|args["samples"]),
chromosomes=args["chromosomes"], num_workers=args["j"], q=args["q"], verbose=args["verbose"])
try:
total = {
sample[1]: sum(total_counts[sample[1], chromosome]
for chromosome in args["chromosomes"])
for sample in args["samples"]
}
total[args["normal"][1]] = sum(total_counts[args["normal"][1],
chromosome]
for chromosome in args["chromosomes"])
except KeyError:
raise KeyError(
"Either a chromosome or a sample has not been considered in the total counting!"
)
log(msg="# Writing the total read counts for all samples in {}\n".format(
args["outputTotal"]),
level="STEP")
with open(args["outputTotal"], 'w') as f:
f.write("{}\t{}\n".format(args["normal"][1], total[args["normal"][1]]))
for sample in args["samples"]:
f.write("{}\t{}\n".format(sample[1], total[sample[1]]))