def open_by_suffix(filename):
if filename.endswith('.gz'):
return gzip.open(filename, 'rt')
elif filename.endswith('.bz2'):
return bz2.BZ2file(filename, 'r')
else: # assume text file
return open(filename, 'r')
def open_by_suffix(filename):
"""Detect compressed file by filename suffix, and return gzip, bz2, or normal file handles."""
if Path(filename).suffix == '.gz':
return gzip.open(filename, 'rt')
elif Path(filename).suffix == '.bz2':
return bz2.BZ2file(filename, 'rt')
else:
return open(filename, 'r')
def open_by_suffix(filename, mode='rt'):
"""Open compressed or uncompressed files."""
if filename.endswith('.gz'):
return gzip.open(filename, mode)
elif filename.endswith('.bz2'):
return bz2.BZ2file(filename, mode)
else:
return open(filename, mode[0])
def phred_parse(fastqlist, sample_size):
phred64dict = {}
counter = 0
for element in fastqlist:
switch = False
print(element)
if element.split('.')[-1] == 'gz' or element.split('.')[-1] == 'gzip':
for line in gzip.open(element, 'r'):
bline = line.decode()[2:-3]
if counter > sample_size: break
elif len(bline) < 3: continue
elif line[0] == "@": continue
else:
switch, counter, breakswitch = do_line(
bline, switch, phred64dict, counter)
if breakswitch == True:
break
elif element.split('.')[-1] == 'bz2' or element.split(
'.')[-1] == 'bzip2':
for line in bz2.BZ2file(element, "r"):
bline = line.decode()[2:-3]
if counter > sample_size: break
elif len(bline) < 3: continue
elif line[0] == "@": continue
else:
switch, counter, breakswitch = do_line(
line, switch, phred64dict, counter)
if breakswitch == True:
break
elif element.split('.')[-1] == 'tar':
for line in tarfile.open(element, "r"):
bline = line.decode()[2:-3]
if counter > sample_size: break
elif len(bline) < 3: continue
elif line[0] == "@": continue
else:
switch, counter, breakswitch = do_line(
line, switch, phred64dict, counter)
if breakswitch == True:
break
else:
for line in open(element):
if counter > sample_size: break
elif len(line) < 3: continue
elif line[0] == "@": continue
else:
switch, counter, breakswitch = do_line(
line, switch, phred64dict, counter)
if breakswitch == True:
break
for element in fastqlist:
if element not in phred64dict:
phred64dict[element] = "33"
return phred64dict
def get_mean_read_len(fastqfile, sample_size, compressed_dict):
mean_read_dict = {}
sampling = []
if (compressed_dict[fastqfile] != "no-compression"
and fastqfile[:fastqfile.rfind(".") + 1][-3:] == "fa") or (
compressed_dict[fastqfile] != "no-compression"
and fastqfile[:fastqfile.rfind(".") + 1][-6:] == ".fasta"):
fastafile = SeqIO.parse(fastqfile, "fasta")
for i in fastafile:
sampling.append(len(i))
if len(sampling) >= sample_size:
break
elif fastqfile[-3:] == "fa" or fastqfile[-6:] == ".fasta":
fastafile = SeqIO.parse(fastqfile, "fasta")
for i in fastafile:
sampling.append(len(i))
if len(sampling) >= sample_size:
break
else:
switch = False
if compressed_dict[fastqfile] == "gzip":
open_fastqfile = gzip.open(fastqfile, 'r')
elif compressed_dict[fastqfile] == "bz2":
open_fastqfile = bz2.BZ2file(fastqfile, "r")
elif compressed_dict[fastqfile] == "tar":
open_fastqfile = tarfile.open(fastqfile, "r")
else:
open_fastqfile = open(fastqfile)
for line in open_fastqfile:
if switch == True:
sampling.append(len(line) - 1)
switch = False
if len(sampling) >= sample_size:
break
if type(line) is str:
if line[0] == "+":
switch = True
elif type(line) == bytes:
try:
bline = line.decode('cp437')
except UnicodeDecodeError:
continue
else:
if bline.find("+") < 5:
switch = True
else:
continue
return [int(numpy.mean(sampling)), numpy.std(sampling)]