def test_calc_gt_type_stats(self):
hdf5 = VariationsH5(join(TEST_DATA_DIR, 'ril.hdf5'), mode='r')
result = calc_gt_type_stats(hdf5)
assert result.shape == (4, 153)
assert numpy.all(numpy.sum(result, axis=0) == 943)
gts = numpy.array([[[0, 0], [1, 1], [0, -1], [-1, -1]],
[[0, -1], [0, 0], [0, -1], [-1, -1]],
[[0, 1], [0, 0], [0, 0], [-1, -1]]])
varis = {'/calls/GT': gts}
res = calc_gt_type_stats(varis)
expected = [[1, 2, 1, 0], [1, 0, 0, 0], [0, 1, 0, 0], [1, 0, 2, 3]]
assert numpy.all(res == expected)
def test_calc_gt_type_stats(self):
hdf5 = VariationsH5(join(TEST_DATA_DIR, 'ril.hdf5'), mode='r')
result = calc_gt_type_stats(hdf5)
assert result.shape == (4, 153)
assert numpy.all(numpy.sum(result, axis=0) == 943)
gts = numpy.array([[[0, 0], [1, 1], [0, -1], [-1, -1]],
[[0, -1], [0, 0], [0, -1], [-1, -1]],
[[0, 1], [0, 0], [0, 0], [-1, -1]]])
varis = {'/calls/GT': gts}
res = calc_gt_type_stats(varis)
expected = [[1, 2, 1, 0], [1, 0, 0, 0], [0, 1, 0, 0], [1, 0, 2, 3]]
assert numpy.all(res == expected)
def plot_gt_stats_per_sample(variations, data_dir, chunk_size=SNPS_PER_CHUNK):
gt_stats = calc_gt_type_stats(variations, chunk_size=chunk_size)
gt_stats = gt_stats.transpose()
figsize = (variations[GT_FIELD].shape[1], 7)
# All genotypes classes per sample
fpath = join(data_dir, 'genotype_counts_per_sample.png')
title = 'Genotypes counts per sample'
mpl_params = {'set_xlabel': {'args': ['Samples'], 'kwargs': {}},
'set_ylabel': {'args': ['Number of GTs'], 'kwargs': {}},
'set_title': {'args': [title], 'kwargs': {}}}
samples = variations.samples
if samples is not None:
mpl_params['set_xticklabels'] = {'args': [samples], 'kwargs': {}}
plot_barplot(gt_stats, ['Ref Homozygous', 'Heterozygous', 'Alt Homozygous',
'Missing GT'], mpl_params=mpl_params,
color=['darkslategrey', 'c', 'paleturquoise', 'cadetblue'],
fpath=fpath, stacked=True, figsize=figsize)
# Missing per sample
fpath = join(data_dir, 'missing_per_sample.png')
title = 'Missing genotypes counts per sample'
mpl_params['set_ylabel'] = {'args': ['Missing Genotypes Number'], 'kwargs': {}}
mpl_params['set_title'] = {'args': [title], 'kwargs': {}}
plot_barplot(gt_stats[:, -1], ['Missing GT'], mpl_params=mpl_params,
fpath=fpath, stacked=True, figsize=figsize)
# Heterozygous per sample
fpath = join(data_dir, 'het_per_sample.png')
title = 'Heterozygous counts per sample'
mpl_params['set_ylabel'] = {'args': ['Heterozygous Number'], 'kwargs': {}}
mpl_params['set_title'] = {'args': [title], 'kwargs': {}}
plot_barplot(gt_stats[:, 1], ['Heterozygous'], mpl_params=mpl_params,
fpath=fpath, stacked=True, figsize=figsize)
# GT percentage without missing values
fpath = join(data_dir, 'gt_perc_per_sample.png')
title = 'Genotypes percentage per sample'
mpl_params['set_ylabel'] = {'args': ['% Genotypes'], 'kwargs': {}}
mpl_params['set_title'] = {'args': [title], 'kwargs': {}}
gt_perc = gt_stats[:, :-1] / gt_stats[:, :-1].sum(axis=1, keepdims=True)
gt_perc *= 100
plot_barplot(gt_perc, ['Ref Homozygous', 'Heterozygous', 'Alt Homozygous'],
mpl_params=mpl_params, fpath=fpath, figsize=figsize)
def plot_gt_stats_per_sample(variations, data_dir, chunk_size=SNPS_PER_CHUNK):
gt_stats = calc_gt_type_stats(variations, chunk_size=chunk_size)
gt_stats = gt_stats.transpose()
figsize = (variations[GT_FIELD].shape[1], 7)
# All genotypes classes per sample
fpath = join(data_dir, 'genotype_counts_per_sample.png')
title = 'Genotypes counts per sample'
mpl_params = {
'set_xlabel': {
'args': ['Samples'],
'kwargs': {}
},
'set_ylabel': {
'args': ['Number of GTs'],
'kwargs': {}
},
'set_title': {
'args': [title],
'kwargs': {}
}
}
samples = variations.samples
if samples is not None:
mpl_params['set_xticklabels'] = {'args': [samples], 'kwargs': {}}
plot_barplot(
gt_stats,
['Ref Homozygous', 'Heterozygous', 'Alt Homozygous', 'Missing GT'],
mpl_params=mpl_params,
color=['darkslategrey', 'c', 'paleturquoise', 'cadetblue'],
fpath=fpath,
stacked=True,
figsize=figsize)
# Missing per sample
fpath = join(data_dir, 'missing_per_sample.png')
title = 'Missing genotypes counts per sample'
mpl_params['set_ylabel'] = {
'args': ['Missing Genotypes Number'],
'kwargs': {}
}
mpl_params['set_title'] = {'args': [title], 'kwargs': {}}
plot_barplot(gt_stats[:, -1], ['Missing GT'],
mpl_params=mpl_params,
fpath=fpath,
stacked=True,
figsize=figsize)
# Heterozygous per sample
fpath = join(data_dir, 'het_per_sample.png')
title = 'Heterozygous counts per sample'
mpl_params['set_ylabel'] = {'args': ['Heterozygous Number'], 'kwargs': {}}
mpl_params['set_title'] = {'args': [title], 'kwargs': {}}
plot_barplot(gt_stats[:, 1], ['Heterozygous'],
mpl_params=mpl_params,
fpath=fpath,
stacked=True,
figsize=figsize)
# GT percentage without missing values
fpath = join(data_dir, 'gt_perc_per_sample.png')
title = 'Genotypes percentage per sample'
mpl_params['set_ylabel'] = {'args': ['% Genotypes'], 'kwargs': {}}
mpl_params['set_title'] = {'args': [title], 'kwargs': {}}
gt_perc = gt_stats[:, :-1] / gt_stats[:, :-1].sum(axis=1, keepdims=True)
gt_perc *= 100
plot_barplot(gt_perc, ['Ref Homozygous', 'Heterozygous', 'Alt Homozygous'],
mpl_params=mpl_params,
fpath=fpath,
figsize=figsize)
