def write_event_durations_to_file(directories, destination):
# make the files to hold durations
template_file = open(destination + "template_durations.csv", 'w')
complement_file = open(destination + "complement_durations.csv", 'w')
all_together_file = open(destination + "all_durations.csv", 'w')
# go through each directory and gather up durations
for directory in directories:
directory_files = directory + "*.fast5"
nb_files = len(glob(directory_files))
sample_set = int(floor(0.25 * nb_files))
print("Found {nb_files} in folder {folder} sampling {nb_sample_set}".format(nb_files=nb_files,
folder=directory,
nb_sample_set=sample_set),
file=sys.stderr)
# make this random.sample, sample 10% of the files in the directory
for f in sample(glob(directory_files), sample_set):
npRead = NanoporeRead(f)
if hasattr(npRead, 'template_event_table') and hasattr(npRead, 'complement_event_table'):
file_name = f.split('/')[-1]
print(file_name, file=sys.stderr)
for t, c in map(None, npRead.template_event_table, npRead.complement_event_table):
if t is not None:
print(t[3], t[6], 't', sep=',', end='\n', file=template_file)
print(t[3], t[6], 't',sep=',', end='\n', file=all_together_file)
if c is not None:
print(c[3], c[6], 'c', sep=',', end='\n', file=complement_file)
print(c[3], c[6], 'c', sep=',', end='\n', file=all_together_file)
npRead.close()
