def peek(sra, outdir=None):
"""return the full file names for all files which will be extracted
Parameters
----------
outdir : path
perform extraction in outdir. If outdir is None, the extraction
will take place in a temporary directory, which will be deleted
afterwards.
Returns
-------
files : list
A list of fastq formatted files that are contained in the archive.
format : string
The quality score format in the :term:`fastq` formatted files.
"""
if outdir is None:
workdir = tempfile.mkdtemp()
else:
workdir = outdir
# --split-files creates files called prefix_#.fastq.gz,
# where # is the read number.
# If file cotains paired end data:
# output = prefix_1.fastq.gz, prefix_2.fastq.gz
# *special case: unpaired reads in a paired end --> prefix.fastq.gz
# *special case: if paired reads are stored in a single read,
# fastq-dump will split. There might be a joining
# sequence. The output would thus be:
# prefix_1.fastq.gz, prefix_2.fastq.gz, prefix_3.fastq.gz
# You want files 1 and 3.
E.run("""fastq-dump --split-files --gzip -X 1000
--outdir %(workdir)s %(sra)s""" % locals())
f = sorted(glob.glob(os.path.join(workdir, "*.fastq.gz")))
ff = [os.path.basename(x) for x in f]
if len(f) == 1:
# sra file contains one read: output = prefix.fastq.gz
pass
elif len(f) == 2:
# sra file contains read pairs:
# output = prefix_1.fastq.gz, prefix_2.fastq.gz
assert ff[0].endswith(
"_1.fastq.gz") and ff[1].endswith("_2.fastq.gz")
elif len(f) == 3:
if ff[2].endswith("_3.fastq.gz"):
f = glob.glob(os.path.join(workdir, "*_[13].fastq.gz"))
else:
f = glob.glob(os.path.join(workdir, "*_[13].fastq.gz"))
# check format of fastqs in .sra
fastq_format = Fastq.guessFormat(IOTools.openFile(f[0], "r"), raises=False)
fastq_datatype = Fastq.guessDataType(
IOTools.openFile(f[0], "r"), raises=True)
if outdir is None:
shutil.rmtree(workdir)
return f, fastq_format, fastq_datatype
def peek(sra, outdir=None):
"""return the full file names for all files which will be extracted
Parameters
----------
outdir : path
perform extraction in outdir. If outdir is None, the extraction
will take place in a temporary directory, which will be deleted
afterwards.
Returns
-------
files : list
A list of fastq formatted files that are contained in the archive.
format : string
The quality score format in the :term:`fastq` formatted files.
"""
if outdir is None:
workdir = tempfile.mkdtemp()
else:
workdir = outdir
# --split-files creates files called prefix_#.fastq.gz,
# where # is the read number.
# If file cotains paired end data:
# output = prefix_1.fastq.gz, prefix_2.fastq.gz
# *special case: unpaired reads in a paired end --> prefix.fastq.gz
# *special case: if paired reads are stored in a single read,
# fastq-dump will split. There might be a joining
# sequence. The output would thus be:
# prefix_1.fastq.gz, prefix_2.fastq.gz, prefix_3.fastq.gz
# You want files 1 and 3.
E.run("""fastq-dump --split-files --gzip -X 1000
--outdir %(workdir)s %(sra)s""" % locals())
f = sorted(glob.glob(os.path.join(workdir, "*.fastq.gz")))
ff = [os.path.basename(x) for x in f]
if len(f) == 1:
# sra file contains one read: output = prefix.fastq.gz
pass
elif len(f) == 2:
# sra file contains read pairs:
# output = prefix_1.fastq.gz, prefix_2.fastq.gz
assert ff[0].endswith("_1.fastq.gz") and ff[1].endswith("_2.fastq.gz")
elif len(f) == 3:
if ff[2].endswith("_3.fastq.gz"):
f = glob.glob(os.path.join(workdir, "*_[13].fastq.gz"))
else:
f = glob.glob(os.path.join(workdir, "*_[13].fastq.gz"))
# check format of fastqs in .sra
fastq_format = Fastq.guessFormat(IOTools.openFile(f[0], "r"), raises=False)
fastq_datatype = Fastq.guessDataType(IOTools.openFile(f[0], "r"),
raises=True)
if outdir is None:
shutil.rmtree(workdir)
return f, fastq_format, fastq_datatype
