def headers(samples, datasets):
'''Statistics headers'''
headers = ['Sample', 'Total reads', 'Mapped reads', 'Deduplicated reads']
splits_headers = set()
for sample in samples:
splits_headers.update([split[len(sample) + 1:] for split in Split.splits(sample)])
if datasets:
for dataset in datasets:
splits_headers.update([split[len(sample) + 1:] for split in Split.splits(sample)])
splits_headers = [header for header in splits_headers]
splits_headers.sort(key=Split.splitkey)
headers.extend(splits_headers)
return (headers, splits_headers)
def sample_splits_prepgenomecov(sample):
'''Prepare BED file used for genome coverage on a single sample.'''
print('Compute genome coverage on sample {}'.format(sample))
prepare_genome_coverage_sample(sample)
splits = sb.splits(sample)
for split in splits:
prepare_genome_coverage_sample(split)
def sample_splits_genome_coverage(sample, genome, scale=None, strand=None, input_suffix='', output_suffix='-cov', spike_suffix=None, control_suffix=None, genomecov_args=()):
'''Compute genome coverage on a single sample.'''
print ('Computing genome coverage on sample {}'.format(sample))
genome_coverage(sample, genome, scale, strand, input_suffix, output_suffix, spike_suffix, control_suffix, genomecov_args)
splits = Split.splits(sample)
for split in splits:
genome_coverage(split, genome, scale, strand, input_suffix, output_suffix, spike_suffix, control_suffix, genomecov_args)
def center_annotations_sample_splits(sample, input_suffix='', output_suffix='-forcov'):
'''Prepare BED file used for genome coverage on a single sample.'''
print ('Center annotations on sample {}'.format(sample))
center_annotations_sample(sample, input_suffix, output_suffix)
splits = sb.splits(sample)
for split in splits:
center_annotations_sample(split, input_suffix, output_suffix)
def dyad_coverage(samples,
genes='genes.txt',
selection=None,
absolute=False,
minp=-75,
maxp=75,
smoothing=None,
suffix=None,
index=None):
'''Finds the distribution of ditances between fragments and dyad.'''
genes_info = pd.read_csv(genes, sep='\t', comment='#')
genes_info = genes_info.loc[genes_info[genes_info.columns[6]] != -1]
if selection:
selection_genes = Parser.first(selection)
genes_info = genes_info[genes_info[genes_info.columns[2]].isin(
selection_genes)]
sample_names = Parser.first(samples)
if index != None:
sample_names = [sample_names[index]]
for sample in sample_names:
dyad_coverage_sample(sample, genes_info, absolute, minp, maxp, suffix,
smoothing)
splits = sb.splits(sample)
for split in splits:
dyad_coverage_sample(split, genes_info, absolute, minp, maxp,
suffix, smoothing)
def fit_gaussian(samples='samples.txt',
absolute=False,
components=False,
svg=False,
verbose=False,
center=None,
cmin=None,
cmax=None,
amp=None,
amin=None,
sigma=None,
smin=None,
suffix=None,
index=None):
'''Fits gaussian curve to dyad coverage.'''
sample_names = Parser.first(samples)
if index != None:
sample_names = [sample_names[index]]
for sample in sample_names:
fit_gaussian_sample(sample, absolute, components, svg, verbose, center,
cmin, cmax, amp, amin, sigma, smin, suffix)
splits = sb.splits(sample)
for split in splits:
fit_gaussian_sample(split, absolute, components, svg, verbose,
center, cmin, cmax, amp, amin, sigma, smin,
suffix)
def ignore_strand_sample_splits(sample,
input_suffix='',
output_suffix='-forcov'):
'''Prepare BED file used for genome coverage on a single sample.'''
print('Duplicate annotations on other strand on sample {}'.format(sample))
ignore_strand_sample(sample, input_suffix, output_suffix)
splits = sb.splits(sample)
for split in splits:
ignore_strand_sample(split, input_suffix, output_suffix)
def test_splits_2(testdir, mock_testclass):
sample = 'POLR2A'
Path(sample + '-100-150.bed').touch()
Path(sample + '-200-250.bed').touch()
Path(sample + '-300-350.bed').touch()
Path(sample + '-400-450.bed').touch()
splits = s.splits(sample)
assert splits[0] == sample + '-100-150'
assert splits[1] == sample + '-200-250'
assert splits[2] == sample + '-300-350'
assert splits[3] == sample + '-400-450'
assert len(splits) == 4
def test_splits(testdir, mock_testclass):
sample = 'POLR2A'
Path(sample + '-100-110.bed').touch()
Path(sample + '-110-120.bed').touch()
Path(sample + '-120-130.bed').touch()
Path(sample + '-130-140.bed').touch()
splits = s.splits(sample)
assert splits[0] == sample + '-100-110'
assert splits[1] == sample + '-110-120'
assert splits[2] == sample + '-120-130'
assert splits[3] == sample + '-130-140'
assert len(splits) == 4
def dyadcov(samples, genes, minp, maxp, smoothing, index):
'''Finds the distribution of ditances between fragments and dyad.'''
logging.basicConfig(filename='debug.log', level=logging.DEBUG, format='%(asctime)s %(levelname)-8s %(message)s', datefmt='%Y-%m-%d %H:%M:%S')
genes_info = pd.read_csv(genes, sep='\t', comment='#')
genes_info = genes_info.loc[genes_info[genes_info.columns[6]] != -1]
sample_names = pd.read_csv(samples, header=None, sep='\t', comment='#')[0]
if index != None:
sample_names = [sample_names[index]]
for sample in sample_names:
dyad_coverage_sample(sample, genes_info, minp, maxp, smoothing)
splits = sb.splits(sample)
for split in splits:
dyad_coverage_sample(split, genes_info, minp, maxp, smoothing)
def fitgaussian(samples, components, svg, verbose, center, cmin, cmax, amp,
amin, sigma, smin, index):
'''Fits gaussian curve to dyad coverage.'''
logging.basicConfig(filename='debug.log',
level=logging.DEBUG,
format='%(asctime)s %(levelname)-8s %(message)s',
datefmt='%Y-%m-%d %H:%M:%S')
sample_names = pd.read_csv(samples, header=None, sep='\t', comment='#')[0]
if index != None:
sample_names = [sample_names[index]]
for sample in sample_names:
fitgaussian_sample(sample, components, svg, verbose, center, cmin,
cmax, amp, amin, sigma, smin)
splits = sb.splits(sample)
for split in splits:
fitgaussian_sample(split, components, svg, verbose, center, cmin,
cmax, amp, amin, sigma, smin)
def fit_double_gaussian(samples='samples.txt',
absolute=False,
components=False,
gaussian=False,
svg=False,
verbose=False,
center1=None,
cmin1=None,
cmax1=None,
amp1=None,
amin1=None,
sigma1=None,
smin1=None,
center2=None,
cmin2=None,
cmax2=None,
amp2=None,
amin2=None,
sigma2=None,
smin2=None,
suffix=None,
index=None):
'''Fits double gaussian curve to dyad coverage.'''
sample_names = Parser.first(samples)
if index != None:
sample_names = [sample_names[index]]
for sample in sample_names:
fit_double_gaussian_sample(sample, absolute, components, gaussian, svg,
verbose, center1, cmin1, cmax1, amp1, amin1,
sigma1, smin1, center2, cmin2, cmax2, amp2,
amin2, sigma2, smin2, suffix)
splits = sb.splits(sample)
for split in splits:
fit_double_gaussian_sample(split, absolute, components, gaussian,
svg, verbose, center1, cmin1, cmax1,
amp1, amin1, sigma1, smin1, center2,
cmin2, cmax2, amp2, amin2, sigma2,
smin2, suffix)
def test_splits_none(testdir, mock_testclass):
sample = 'POLR2A'
splits = s.splits(sample)
assert len(splits) == 0