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_calc_allele_obs_gq_distrib_2D(self):
variations = {'/calls/AO': numpy.array([[[0, 0], [5, 0], [-1, -1],
[0, -1], [0, 0], [0, 10],
[20, 0], [25, 0], [20, 20],
[0, 0]]]),
'/calls/RO': numpy.array([[0, 5, 15, 7, 10, 0, 0, 25,
20, 10]]),
'/calls/GQ': numpy.array([[40, 30, 35, 30, 0,
40, 30, 35, 30, 0]]),
'/calls/GT': numpy.array([[[0, 0], [1, 0], [-1, -1],
[0, -1], [0, 0], [0, 10],
[1, 0], [0, 0], [0, 0],
[1, 0]]])}
hist, _, _ = hist2d_gq_allele_observations(variations, chunk_size=None)
assert hist[0, 0] == 40
assert hist[-1, -1] == 35
hdf5 = VariationsH5(join(TEST_DATA_DIR, 'ril.hdf5'), mode='r')
hist, xbins, ybins = hist2d_gq_allele_observations(hdf5)
hist2, xbins2, ybins2 = hist2d_gq_allele_observations(hdf5,
chunk_size=10)
assert numpy.allclose(xbins, xbins2)
assert numpy.allclose(ybins, ybins2)
assert numpy.all(hist == hist2)
def test_calc_allele_obs_gq_distrib_2D(self):
variations = {'/calls/AO': numpy.array([[[0, 0], [5, 0], [-1, -1],
[0, -1], [0, 0], [0, 10],
[20, 0], [25, 0], [20, 20],
[0, 0]]]),
'/calls/RO': numpy.array([[0, 5, 15, 7, 10, 0, 0, 25,
20, 10]]),
'/calls/GQ': numpy.array([[40, 30, 35, 30, 0,
40, 30, 35, 30, 0]]),
'/calls/GT': numpy.array([[[0, 0], [1, 0], [-1, -1],
[0, -1], [0, 0], [0, 10],
[1, 0], [0, 0], [0, 0],
[1, 0]]])}
hist, _, _ = hist2d_gq_allele_observations(variations, chunk_size=None)
assert hist[0, 0] == 40
assert hist[-1, -1] == 35
hdf5 = VariationsH5(join(TEST_DATA_DIR, 'ril.hdf5'), mode='r')
hist, xbins, ybins = hist2d_gq_allele_observations(hdf5)
hist2, xbins2, ybins2 = hist2d_gq_allele_observations(hdf5,
chunk_size=10)
assert numpy.allclose(xbins, xbins2)
assert numpy.allclose(ybins, ybins2)
assert numpy.all(hist == hist2)
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)