def get_coverage_files_helper(filename, region_intervals_file):
# First makes the three bed file
three_prime_end_file = make_read_end_file(filename, 'three')
# Run coverage on the three bed file
coverage_file = generate_random_filename()
run_coverage(region_intervals_file,
three_prime_end_file,
output_filename=coverage_file)
remove_files(three_prime_end_file)
return coverage_file
def get_counts_in_gene_bodies(regions_filename, blacklisted_sequencing_file,
indv_gene_counts_dict):
three_bed_filename = make_read_end_file(blacklisted_sequencing_file,
'three')
random_filename = run_coverage(regions_filename, three_bed_filename)
with open(random_filename) as file:
for line in file:
counts = int(line.split()[-4])
gene_name = line.split()[3]
indv_gene_counts_dict[gene_name]["Body"] = counts
remove_files(three_bed_filename, random_filename)
return indv_gene_counts_dict
def quantify_intervals(sequencing_filename, blacklist_filename,
intervals_file):
# First make the sequencing file 3' ends only
three_prime_filename = make_read_end_file(sequencing_filename, 'three')
# Now blacklist the sequencing file
blacklisted_sequencing_filename = run_subtract(three_prime_filename,
blacklist_filename,
rna_blacklist_file)
# Use bedtools coverage to get the number of 3' reads for each interval
coverage_file = run_coverage(intervals_file,
blacklisted_sequencing_filename)
# We can now remove the 3' reads file
remove_files(blacklisted_sequencing_filename, three_prime_filename)
return coverage_file
def gather_data(read_type, seq_file, regions_file):
if read_type in ['five', 'three']:
modified_seq_filename = make_read_end_file(seq_file, read_type)
need_to_remove_modified_seq_filename = True
else:
modified_seq_filename = seq_file
need_to_remove_modified_seq_filename = False
# Then quantify the 5' ends
coverage_file = run_coverage(regions_file, modified_seq_filename)
data = organize_counts(coverage_file, seq_file)
remove_files(coverage_file)
if need_to_remove_modified_seq_filename:
remove_files(modified_seq_filename)
return data
def get_counts_in_paused_region(pause_region_filename,
blacklisted_sequencing_file):
five_bed_filename = make_read_end_file(blacklisted_sequencing_file, 'five')
# Run bedtools coverage on the 5' bed file
random_filename = run_coverage(pause_region_filename, five_bed_filename)
# Not using a defaultdict because multiprocessing does not like it
indv_gene_counts_dict = {}
with open(random_filename) as file:
for line in file:
counts = int(line.split()[-4])
gene_name = line.split()[3]
if gene_name not in indv_gene_counts_dict:
indv_gene_counts_dict[gene_name] = {"Pause": -1, "Body": -1}
indv_gene_counts_dict[gene_name]["Pause"] = counts
remove_files(five_bed_filename, random_filename)
return indv_gene_counts_dict