def vcf_dp(fn_list, max_dp=1024):
dp_array = np.zeros(max_dp, dtype=int)
qual_dictlist = collections.defaultdict(list)
for filename in fn_list:
vp = vcf_parser(filename, yield_samples=True, parse_genotypes=True)
samples = next(vp)
if len(samples) != 1:
raise Exception("{} samples. Single sample vcf expected.".format(len(samples)))
s = samples[0]
for vline in vp:
qual = float(vline[7])
genotype_field = vline[11][s]
ad = genotype_field["AD"]
dp = np.sum(map(int, ad.split(",")))
if dp < max_dp:
dp_array[dp] += 1
qual_dictlist[dp].append(qual)
qual_mean = [np.mean(qual_dictlist[i]) if len(qual_dictlist[i])>0 else 0 for i in range(max_dp)]
qual_var = [np.var(qual_dictlist[i]) if len(qual_dictlist[i])>0 else 0 for i in range(max_dp)]
return dp_array, qual_mean, qual_var
def filter_snps_by_dist(vcf_filename, plen):
"""
Yield SNPs which are plen positions away from other SNPs.
"""
vp = vcf_parser(vcf_filename, yield_samples=True, parse_genotypes=True)
samples = next(vp)
if len(samples) != 1:
raise Exception("{} samples. Single sample vcf expected.".format(
len(samples)))
sample = samples[0]
last = None
exclude = False
for vline in vp:
chrom = vline[2]
pos = vline[3]
ref = vline[5].upper()
alt = vline[6].upper()
genotype_field = vline[11][sample]
ad = genotype_field["AD"]
dp = np.sum(map(int, ad.split(",")))
if last is not None:
if last[0] == chrom and pos - last[1] < plen:
exclude = True
elif not exclude:
yield last
else:
exclude = False
last = [chrom, pos, ref, alt, dp]
if not exclude:
yield last
if __name__ == "__main__":
args = parse_args()
chrom_dict = parse_fai(args.fai)
loci = parse_bed(args.bed)
total_dp = collections.defaultdict(lambda: collections.Counter())
pdf = PdfPages("plot_chrom_cov_from_vcf.pdf")
fig_w, fig_h = plt.figaspect(4.0 / 1.0)
#fig_w, fig_h = plt.figaspect(9.0/16.0)
#fig_w, fig_h = plt.figaspect(3.0/4.0)
for filename in args.vcf:
vparser = parsers.vcf_parser(filename)
samples = next(vparser)
x_pos = collections.defaultdict(lambda: collections.defaultdict(list))
y_dp = collections.defaultdict(lambda: collections.defaultdict(list))
max_dp = collections.defaultdict(int)
for row in vparser:
(lineno, line, chrom, pos, id, ref, alt, qual, filter, info_str,
fmt_fields, gts) = row
for sample in samples:
dp = gts[sample].get("DP")
if not dp:
print(
from __future__ import print_function
import sys
from parsers import vcf_parser
import numpy as np
import collections
if __name__ == "__main__":
if len(sys.argv) < 3:
print("usage: {} file.vcf DP1 [...DPN]".format(sys.argv[0]),
file=sys.stderr)
exit(1)
dp_match = set(map(int, sys.argv[2:]))
vp = vcf_parser(sys.argv[1], yield_samples=True, parse_genotypes=True)
samples = next(vp)
if len(samples) != 1:
raise Exception("{} samples. Single sample vcf expected.".format(
len(samples)))
s = samples[0]
for vline in vp:
chrom = vline[2]
pos = vline[3]
qual = float(vline[7])
genotype_field = vline[11][s]
ad = genotype_field["AD"]
dp = np.sum(map(int, ad.split(",")))
if dp in dp_match:
import matplotlib.gridspec as gridspec
import parsers
if __name__ == "__main__":
if len(sys.argv) < 2:
print("usage: {} in1.vcf [in2.vcf...inN.vcf]".format(sys.argv[0]))
sys.exit(1)
pdf = PdfPages("plot_hist_dp_qual.pdf")
fig_w, fig_h = plt.figaspect(9.0 / 16.0)
#fig_w, fig_h = plt.figaspect(3.0/4.0)
for vcf_in in sys.argv[1:]:
vparser = parsers.vcf_parser(vcf_in)
samples = next(vparser)
if (len(samples) != 1):
print("Error: {}: multisample vcf is unexpected.".format(vcf_in),
file=sys.stderr)
sys.exit(1)
sample = samples[0]
qual_list = []
dp_list = []
gt_count = collections.Counter()
for row in vparser:
(lineno, line, chrom, pos, id, ref, alt, qual, filter, info_str,
fmt_fields, gts) = row
print("Error: no individuals in group ``{}''".format(args.group),
file=sys.stderr)
exit(1)
else:
ii = list(range(len(indlist)))
alleles = {}
for (chrom, pos), g in zip(parse_snp(snp_fn), parse_geno(geno_fn)):
gts = [int(g[i]) for i in ii if g[i] != "9"]
if len(gts) == 0:
continue
n_ref = sum(gts)
n_alt = 2 * len(gts) - n_ref
alleles[(chrom, pos)] = (n_alt, n_ref + n_alt)
vparser = vcf_parser(args.vcf)
vcf_samples = next(vparser)
assert len(vcf_samples) == 1, "multi-sample vcf not supported"
sample = vcf_samples[0]
gt_map = {"./.": -1, ".|.": -1}
for a0 in range(9):
for a1 in range(9):
n_alts = sum(1 if a else 0 for a in [a0, a1])
gt_map["{}/{}".format(a0, a1)] = n_alts
gt_map["{}|{}".format(a0, a1)] = n_alts
prev_pos = 0
prev_chrom = None
last = (-1, -1, -1)