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 statistics_samples(samples='samples.txt', datasets='dataset.txt', output='statistics.txt'):
'''Creates statistics file for samples.'''
sample_names = Parser.first(samples)
datasets_names = []
if os.path.exists(datasets):
datasets_names = Parser.first(datasets)
compute_statistics(sample_names, datasets_names, output)
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 removesecondmate_samples(samples='samples.txt', input_suffix='-dedup', output_suffix='-mate1', threads=None, index=None):
'''Removes second mate from BAM.'''
sample_names = Parser.first(samples)
if index != None:
sample_names = [sample_names[index]]
for sample in sample_names:
removesecondmate_sample(sample, input_suffix, output_suffix, threads)
def split_samples(samples='samples.txt', index=None, binlength=10, binminlength=100, binmaxlength=500):
'''Split BED files from samples based on lenght of annotations.'''
sample_names = Parser.first(samples)
if index != None:
sample_names = [sample_names[index]]
for sample in sample_names:
split_sample(sample, binlength, binminlength, binmaxlength)
def genome_coverage_samples(samples='samples.txt', genome='sacCer3.chrom.sizes', scale=None, strand=None, input_suffix='', output_suffix='-cov', spike_suffix=None, control_suffix=None, index=None, genomecov_args=()):
'''Compute genome coverage on samples.'''
sample_names = Parser.first(samples)
if index != None:
sample_names = [sample_names[index]]
for sample in sample_names:
sample_splits_genome_coverage(sample, genome, scale, strand, input_suffix, output_suffix, spike_suffix, control_suffix, genomecov_args)
def center_annotations_samples(samples='samples.txt', input_suffix='', output_suffix='-forcov', index=None):
'''Prepare BED file used for genome coverage on samples.'''
sample_names = Parser.first(samples)
if index != None:
sample_names = [sample_names[index]]
for sample in sample_names:
center_annotations_sample_splits(sample, input_suffix, output_suffix)
def intersectannotations(input, annotations, output):
'''Filter BED file to keep only annotations present in annotations.'''
logging.basicConfig(filename='seqtools.log',
level=logging.DEBUG,
format='%(asctime)s %(levelname)-8s %(message)s',
datefmt='%Y-%m-%d %H:%M:%S')
genes = Parser.first(annotations)
incolumns = Parser.columns(input)
with open(output, 'w') as outfile:
for columns in incolumns:
if columns[3] in genes:
outfile.write(str(columns[0]))
for column in columns[1:]:
outfile.write('\t')
outfile.write(str(column))
outfile.write('\n')
def plot2do_samples(file, input_suffix='', index=None, plot2do_args=()):
'''Run plot2DO on samples.'''
file_parent = Path(file).parent
sample_names = Parser.first(file)
if index != None:
sample_names = [sample_names[index]]
for sample in sample_names:
plot2do_sample(str(file_parent / sample), input_suffix, plot2do_args)
def test_columns():
samples = Path(__file__).parent.parent.joinpath('samples.txt')
columns = p.columns(samples)
assert columns[0][0] == 'POLR2A'
assert columns[0][1] == 'SRR8518913'
assert columns[1][0] == 'ASDURF'
assert columns[1][1] == 'SRX5322424'
assert columns[2][0] == 'POLR1C'
assert columns[2][1] == 'SRR8518915'
def merge_datasets(datasets='dataset.txt', index=None):
'''Merge BED files related to samples.'''
datasets_columns = Parser.columns(datasets)
if index != None:
datasets_columns = [datasets_columns[index]]
for columns in datasets_columns:
name = columns[0]
samples = [sample for sample in columns[1:]]
merge_dataset(name, samples)
def bowtie_samples(samples='samples.txt',
threads=None,
output_suffix='',
index=None,
bowtie_args=()):
'''Align samples using bowtie2 program.'''
sample_names = Parser.first(samples)
if index != None:
sample_names = [sample_names[index]]
for sample in sample_names:
bowtie_sample(sample, threads, output_suffix, bowtie_args)
def merge_datasets(datasets='dataset.txt',
sizes='sacCer3.chrom.sizes',
index=None):
'''Merge bigWig files related to samples.'''
datasets_columns = Parser.columns(datasets)
if index != None:
datasets_columns = [datasets_columns[index]]
for columns in datasets_columns:
name = columns[0]
samples = [sample for sample in columns[1:]]
merge_dataset(name, samples, sizes)
def bwa_samples(samples='samples.txt',
fasta='sacCer3.fa',
threads=None,
output_suffix='',
index=None,
bwa_args=()):
'''Align samples using bwa program.'''
sample_names = Parser.first(samples)
if index != None:
sample_names = [sample_names[index]]
for sample in sample_names:
bwa_sample(sample, fasta, threads, output_suffix, bwa_args)
def shift_annotations_samples(samples='samples.txt',
input_suffix='',
output_suffix='-forcov',
index=None,
bedtools_args=()):
'''Moves annotations contained in BED files.'''
sample_names = Parser.first(samples)
if index != None:
sample_names = [sample_names[index]]
for sample in sample_names:
shift_annotations_sample(sample, input_suffix, output_suffix,
bedtools_args)
def test_columns_merge():
samples = Path(__file__).parent.parent.joinpath('dataset.txt')
columns = p.columns(samples)
assert columns[0][0] == 'POLR2A'
assert columns[0][1] == 'POLR2A_1'
assert columns[0][2] == 'POLR2A_2'
assert columns[1][0] == 'ASDURF'
assert columns[1][1] == 'ASDURF_1'
assert columns[1][2] == 'ASDURF_2'
assert columns[2][0] == 'POLR1C'
assert columns[2][1] == 'POLR1C_1'
assert columns[2][2] == 'POLR1C_2'
def download_samples(samples='samples.txt',
fast=True,
threads=None,
mem='100MB',
index=None):
'''Download reads of all samples.'''
sample_columns = Parser.columns(samples)
if index != None:
sample_columns = [sample_columns[index]]
for columns in sample_columns:
sample = columns[0]
srr = columns[1] if len(columns) > 1 else None
download_sample(sample, srr, fast, threads, mem)
def filter_bam(samples='samples.txt',
paired=True,
dedup=True,
threads=None,
input_suffix='',
output_suffix='',
index=None):
'''Filter BAM file to keep only properly paired reads and remove supplementary alignments and duplicates.'''
sample_names = Parser.first(samples)
if index != None:
sample_names = [sample_names[index]]
for sample in sample_names:
filter_bam_sample(sample, paired, dedup, threads, input_suffix,
output_suffix)
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 merge_dataset(name, samples, sizes):
'''Merge bigWig files related to samples.'''
print('Merging samples {} into dataset {}'.format(samples, name))
sizes_columns = Parser.columns(sizes)
bws = [pbw.open(sample + '.bw') for sample in samples]
merge_temp_o, merge_temp = tempfile.mkstemp(suffix='.bed')
with open(merge_temp_o, 'w') as output:
output.write('track type=bedGraph name="' + name + '"\n')
for size_columns in sizes_columns:
chromosome = size_columns[0]
size = size_columns[1]
sums = [0] * size
for bw in bws:
bw_size = bw.chroms(chromosome) if bw.chroms(chromosome) else 0
if bw_size == 0:
continue
values = bw.values(chromosome, 0, min(size, bw_size))
sums = [
sums[i] + (values[i] if not math.isnan(values[i]) else 0)
for i in range(0, min(size, bw_size))
]
for i in range(0, len(sums)):
output.write(chromosome)
output.write('\t')
output.write(str(i))
output.write('\t')
output.write(str(i + 1))
output.write('\t')
output.write(str(sums[i]))
output.write('\n')
sort_temp_o, sort_temp = tempfile.mkstemp(suffix='.bed')
Bed.sort(merge_temp, sort_temp)
merged_bw = name + '.bw'
Bed.bedgraph_to_bigwig(sort_temp, merged_bw, sizes)
os.remove(sort_temp)
os.remove(merge_temp)
def test_first_merge():
samples = Path(__file__).parent.parent.joinpath('dataset.txt')
names = p.first(samples)
assert names == ['POLR2A', 'ASDURF', 'POLR1C']