def plot_allele_obs_distrib_2D(variations, data_dir, max_allele_counts,
chunk_size=SNPS_PER_CHUNK):
# Allele observation distribution 2D
masks = [call_is_het, call_is_hom_alt, call_is_hom_ref]
names = ['Heterozygous', 'Alt Homozygous', 'Ref Homozygous']
fig = Figure(figsize=(22, 25))
canvas = FigureCanvas(fig)
gs = gridspec.GridSpec(3, 2)
fpath = join(data_dir, 'allele_obs_distrib_per_gt.png')
fhand = open(fpath, 'w')
counts_range = [[0, max_allele_counts], [0, max_allele_counts]]
for i, (mask_func, name) in enumerate(zip(masks, names)):
hist2d = hist2d_allele_observations(variations,
n_bins=max_allele_counts,
range_=counts_range,
mask_func=mask_func,
chunk_size=chunk_size)
counts_distrib2d, xbins, ybins = hist2d
axes = fig.add_subplot(gs[i, 0])
title = 'Allele counts distribution 2D {}'.format(name)
plot_hist2d(numpy.log10(counts_distrib2d), xbins, ybins, axes=axes,
mpl_params={'set_xlabel': {'args': ['Alt allele counts'],
'kwargs': {}},
'set_ylabel': {'args': ['Ref allele counts'],
'kwargs': {}},
'set_title': {'args': [title], 'kwargs': {}}},
colorbar_label='log10(counts)', fig=fig)
hist2d = hist2d_gq_allele_observations(variations,
n_bins=max_allele_counts,
range_=counts_range,
mask_func=mask_func,
chunk_size=chunk_size,
hist_counts=counts_distrib2d)
gq_distrib2d, xbins, ybins = hist2d
axes = fig.add_subplot(gs[i, 1])
title = 'Allele counts GQ distribution 2D {}'.format(name)
plot_hist2d(gq_distrib2d, xbins, ybins, axes=axes, fig=fig,
mpl_params={'set_xlabel': {'args': ['Alt allele counts'],
'kwargs': {}},
'set_ylabel': {'args': ['Ref allele counts'],
'kwargs': {}},
'set_title': {'args': [title], 'kwargs': {}}},
colorbar_label='Genotype Quality (GQ)')
canvas.print_figure(fhand)
def test_plot_hist2d(self):
x = numpy.random.normal(loc=1, size=1000)
y = numpy.random.normal(size=1000)
distrib, xbins, ybins = numpy.histogram2d(x, y)
with NamedTemporaryFile(suffix='.png') as fhand:
plot_hist2d(distrib,
xbins,
ybins,
fhand=fhand,
colorbar_label='Counts',
hist1d=False)
with NamedTemporaryFile(suffix='.png') as fhand:
plot_hist2d(distrib,
xbins,
ybins,
fhand=fhand,
colorbar_label='Counts')
with NamedTemporaryFile(suffix='.png') as fhand:
plot_hist2d(distrib,
xbins,
ybins,
fhand=fhand,
log_normed=True,
mpl_params={
'set_xlabel': 'x',
'set_ylabel': 'y'
})
def test_plot_hist2d(self):
x = numpy.random.normal(loc=1, size=1000)
y = numpy.random.normal(size=1000)
distrib, xbins, ybins = numpy.histogram2d(x, y)
with NamedTemporaryFile(suffix=".png") as fhand:
plot_hist2d(distrib, xbins, ybins, fhand=fhand, colorbar_label="Counts", hist1d=False)
with NamedTemporaryFile(suffix=".png") as fhand:
plot_hist2d(distrib, xbins, ybins, fhand=fhand, colorbar_label="Counts")
with NamedTemporaryFile(suffix=".png") as fhand:
plot_hist2d(
distrib, xbins, ybins, fhand=fhand, log_normed=True, mpl_params={"set_xlabel": "x", "set_ylabel": "y"}
)
def plot_allele_obs_distrib_2D(variations,
data_dir,
max_allele_counts,
chunk_size=SNPS_PER_CHUNK):
# Allele observation distribution 2D
masks = [call_is_het, call_is_hom_alt, call_is_hom_ref]
names = ['Heterozygous', 'Alt Homozygous', 'Ref Homozygous']
fig = Figure(figsize=(22, 25))
canvas = FigureCanvas(fig)
gs = gridspec.GridSpec(3, 2)
fpath = join(data_dir, 'allele_obs_distrib_per_gt.png')
fhand = open(fpath, 'w')
counts_range = [[0, max_allele_counts], [0, max_allele_counts]]
for i, (mask_func, name) in enumerate(zip(masks, names)):
hist2d = hist2d_allele_observations(variations,
n_bins=max_allele_counts,
range_=counts_range,
mask_func=mask_func,
chunk_size=chunk_size)
counts_distrib2d, xbins, ybins = hist2d
axes = fig.add_subplot(gs[i, 0])
title = 'Allele counts distribution 2D {}'.format(name)
plot_hist2d(numpy.log10(counts_distrib2d),
xbins,
ybins,
axes=axes,
mpl_params={
'set_xlabel': {
'args': ['Alt allele counts'],
'kwargs': {}
},
'set_ylabel': {
'args': ['Ref allele counts'],
'kwargs': {}
},
'set_title': {
'args': [title],
'kwargs': {}
}
},
colorbar_label='log10(counts)',
fig=fig)
hist2d = hist2d_gq_allele_observations(variations,
n_bins=max_allele_counts,
range_=counts_range,
mask_func=mask_func,
chunk_size=chunk_size,
hist_counts=counts_distrib2d)
gq_distrib2d, xbins, ybins = hist2d
axes = fig.add_subplot(gs[i, 1])
title = 'Allele counts GQ distribution 2D {}'.format(name)
plot_hist2d(gq_distrib2d,
xbins,
ybins,
axes=axes,
fig=fig,
mpl_params={
'set_xlabel': {
'args': ['Alt allele counts'],
'kwargs': {}
},
'set_ylabel': {
'args': ['Ref allele counts'],
'kwargs': {}
},
'set_title': {
'args': [title],
'kwargs': {}
}
},
colorbar_label='Genotype Quality (GQ)')
canvas.print_figure(fhand)
