def __init__(self, fhand, compressed=None, filename=None,
min_calls_for_pop_stats=DEF_MIN_CALLS_FOR_POP_STATS):
self.fhand = fhand
self.pyvcf_reader = pyvcfReader(fsock=fhand, compressed=compressed,
filename=filename)
self.min_calls_for_pop_stats = min_calls_for_pop_stats
self._snpcaller = None
def __init__(self, vcf_fhand, parents_a, parents_b,
threshold=DEF_AB_CODER_THRESHOLD,
offspring=None, window=DEF_AB_CODING_WIN):
self._reader = pyvcfReader(vcf_fhand)
self.parents_a = parents_a
self.parents_b = parents_b
self._offspring = offspring
self.window = window
self.threshold = threshold
self.log = Counter()
self.indexes = array('f')
def calc_recomb_rates_along_chroms(vcf_fpath, pop_type, samples=None,
min_phys_dist=1000, max_phys_dist=50000):
reader = pyvcfReader(open(vcf_fpath))
random_reader = pyvcfReader(open(vcf_fpath))
diffs_from_r_expected = []
diffs_distort =[]
recomb_rates = []
diffs_xs = []
diffs_ys = []
to_return = []
r_xs = []
r_ys = []
r_dist_color = []
#samples = reader.samples
#samples = [sample for sample in samples if sample[0] == '4']
for snv in reader:
pos = snv.POS
chrom = snv.CHROM
start = pos - max_phys_dist
if start < 0:
start = 0
end = pos - min_phys_dist
if end < 0:
end = 0
if end > 0:
snvs_prev = random_reader.fetch(chrom=chrom, start=start, end=end)
else:
snvs_prev = iter([])
start = pos + min_phys_dist
end = pos + max_phys_dist
snvs_after = random_reader.fetch(chrom=chrom, start=start, end=end)
if samples is None:
calls1 = snv.samples
else:
calls1 = [call for call in snv.samples if call.sample in samples]
recombs = []
for snv_2 in chain(snvs_prev, snvs_after):
dist = abs(pos - snv_2.POS)
if samples is None:
calls2 = snv_2.samples
else:
calls2 = [call for call in snv_2.samples if call.sample in samples]
result = _calc_recomb_rate(calls1, calls2, pop_type)
if result:
recomb, haplo_count = result
else:
recomb, haplo_count = None, None
if haplo_count:
diff_from_r_expected = ((abs(haplo_count.AB - haplo_count.ab) +
abs(haplo_count.Ab - haplo_count.aB)) /
sum(haplo_count))
diff_distort = ((abs((haplo_count.Ab + haplo_count.AB) - (haplo_count.ab + haplo_count.aB)) +
abs((haplo_count.AB + haplo_count.aB) - (haplo_count.ab + haplo_count.Ab))) /
sum(haplo_count))
diffs_xs.append(diff_from_r_expected)
diffs_ys.append(diff_distort)
diffs_from_r_expected.append(diff_from_r_expected)
diffs_distort.append(diff_distort)
if recomb is not None:
recomb_rates.append(recomb)
r_xs.append(recomb)
r_ys.append(diff_from_r_expected)
r_dist_color.append(diff_distort)
#if 0.6 > recomb > 0.2 and diff_from_r_expected > 0.9:
# print 'pau->', haplo_count, recomb, diff_from_r_expected
#if recomb > 0.5:
if recomb is None:
continue
print snv.CHROM, snv.POS, snv_2.CHROM, snv_2.POS, recomb, diff_from_r_expected, diff_distort
recombs.append((dist, recomb))
if recombs:
to_return.append([chrom, pos, recombs])
from crumbs.plot import build_histogram, draw_density_plot
print 'hola'
from os.path import join as pjoin
dir_ = '/home/jose/tmp/rils/results/'
fpath = pjoin(dir_, 'diffs_from_r.png')
build_histogram(diffs_from_r_expected, open(fpath, 'w'), bins=30)
fpath = pjoin(dir_, 'diffs_distort.png')
build_histogram(diffs_distort, open(fpath, 'w'), bins=30)
fpath = pjoin(dir_, 'recomb_rates.png')
build_histogram(recomb_rates, open(fpath, 'w'), bins=30, log=True)
fpath = pjoin(dir_, 'diffs_density.png')
draw_density_plot(diffs_xs, diffs_ys, open(fpath, 'w'))
fpath = pjoin(dir_, 'diffs_scatter.png')
draw_scatter(diffs_xs, diffs_ys, open(fpath, 'w'))
fpath = pjoin(dir_, 'r_vs_diff_r_scatter.png')
print len(r_xs), len(r_ys)
draw_scatter(r_xs, r_ys, open(fpath, 'w'), color=r_dist_color)
fpath = pjoin(dir_, 'r_vs_diff_r_density.png')
draw_density_plot(r_xs, r_ys, open(fpath, 'w'))
return to_return