def align_transcriptome(fastq_file, pair_file, ref_file, data):
"""
bwa mem with settings for aligning to the transcriptome for eXpress/RSEM/etc
"""
work_bam = dd.get_work_bam(data)
base, ext = os.path.splitext(work_bam)
out_file = base + ".transcriptome" + ext
if utils.file_exists(out_file):
data = dd.set_transcriptome_bam(data, out_file)
return data
# bwa mem needs phred+33 quality, so convert if it is Illumina
if dd.get_quality_format(data).lower() == "illumina":
logger.info("bwa mem does not support the phred+64 quality format, "
"converting %s and %s to phred+33.")
fastq_file = fastq.groom(fastq_file, data, in_qual="fastq-illumina")
if pair_file:
pair_file = fastq.groom(pair_file, data, in_qual="fastq-illumina")
bwa = config_utils.get_program("bwa", data["config"])
gtf_file = dd.get_gtf_file(data)
gtf_fasta = index_transcriptome(gtf_file, ref_file, data)
args = " ".join(_bwa_args_from_config(data["config"]))
num_cores = data["config"]["algorithm"].get("num_cores", 1)
samtools = config_utils.get_program("samtools", data["config"])
cmd = ("{bwa} mem {args} -a -t {num_cores} {gtf_fasta} {fastq_file} "
"{pair_file} | {samtools} view -bhS - > {tx_out_file}")
with file_transaction(data, out_file) as tx_out_file:
message = "Aligning %s and %s to the transcriptome." % (fastq_file,
pair_file)
do.run(cmd.format(**locals()), message)
data = dd.set_transcriptome_bam(data, out_file)
return data
def align_transcriptome(fastq_file, pair_file, ref_file, data):
"""
bwa mem with settings for aligning to the transcriptome for eXpress/RSEM/etc
"""
work_bam = dd.get_work_bam(data)
base, ext = os.path.splitext(work_bam)
out_file = base + ".transcriptome" + ext
if utils.file_exists(out_file):
data = dd.set_transcriptome_bam(data, out_file)
return data
# bwa mem needs phred+33 quality, so convert if it is Illumina
if dd.get_quality_format(data).lower() == "illumina":
logger.info("bwa mem does not support the phred+64 quality format, " "converting %s and %s to phred+33.")
fastq_file = fastq.groom(fastq_file, in_qual="fastq-illumina", data=data)
if pair_file:
pair_file = fastq.groom(pair_file, in_qual="fastq-illumina", data=data)
bwa = config_utils.get_program("bwa", data["config"])
gtf_file = dd.get_gtf_file(data)
gtf_fasta = index_transcriptome(gtf_file, ref_file, data)
args = " ".join(_bwa_args_from_config(data["config"]))
num_cores = data["config"]["algorithm"].get("num_cores", 1)
cmd = (
"{bwa} mem {args} -a -t {num_cores} {gtf_fasta} {fastq_file} "
"{pair_file} | samtools view -bhS - > {tx_out_file}"
)
with file_transaction(out_file) as tx_out_file:
message = "Aligning %s and %s to the transcriptome." % (fastq_file, pair_file)
do.run(cmd.format(**locals()), message)
data = dd.set_transcriptome_bam(data, out_file)
return data
def get_fastq_files(data):
"""Retrieve fastq files for the given lane, ready to process.
"""
assert "files" in data, "Did not find `files` in input; nothing to process"
ready_files = []
should_gzip = True
# Bowtie does not accept gzipped fastq
if 'bowtie' in data['reference'].keys():
should_gzip = False
for fname in data["files"]:
if fname.endswith(".bam"):
if _pipeline_needs_fastq(data["config"], data):
ready_files = convert_bam_to_fastq(fname, data["dirs"]["work"],
data, data["dirs"], data["config"])
else:
ready_files = [fname]
elif objectstore.is_remote(fname):
ready_files.append(fname)
# Trimming does quality conversion, so if not doing that, do an explicit conversion
elif not(dd.get_trim_reads(data)) and dd.get_quality_format(data) != "standard":
out_dir = utils.safe_makedir(os.path.join(dd.get_work_dir(data), "fastq_convert"))
ready_files.append(fastq.groom(fname, data, out_dir=out_dir))
else:
ready_files.append(fname)
ready_files = [x for x in ready_files if x is not None]
if should_gzip:
out_dir = utils.safe_makedir(os.path.join(dd.get_work_dir(data), "fastq"))
ready_files = [_gzip_fastq(x, out_dir) for x in ready_files]
for in_file in ready_files:
if not objectstore.is_remote(in_file):
assert os.path.exists(in_file), "%s does not exist." % in_file
return ready_files
def get_fastq_files(data):
"""Retrieve fastq files for the given lane, ready to process.
"""
assert "files" in data, "Did not find `files` in input; nothing to process"
ready_files = []
should_gzip = True
# Bowtie does not accept gzipped fastq
if 'bowtie' in data['reference'].keys():
should_gzip = False
for fname in data["files"]:
if fname.endswith(".bam"):
if _pipeline_needs_fastq(data["config"], data):
ready_files = convert_bam_to_fastq(fname, data["dirs"]["work"],
data, data["dirs"], data["config"])
else:
ready_files = [fname]
elif objectstore.is_remote(fname):
ready_files.append(fname)
# Trimming does quality conversion, so if not doing that, do an explicit conversion
elif not(dd.get_trim_reads(data)) and dd.get_quality_format(data) != "standard":
out_dir = utils.safe_makedir(os.path.join(dd.get_work_dir(data), "fastq_convert"))
ready_files.append(fastq.groom(fname, data, out_dir=out_dir))
else:
ready_files.append(fname)
ready_files = [x for x in ready_files if x is not None]
if should_gzip:
out_dir = utils.safe_makedir(os.path.join(dd.get_work_dir(data), "fastq"))
ready_files = [_gzip_fastq(x, out_dir) for x in ready_files]
for in_file in ready_files:
if not objectstore.is_remote(in_file):
assert os.path.exists(in_file), "%s does not exist." % in_file
return ready_files
def test_groom(self):
illumina_dir = os.path.join(self.root_dir, "illumina")
test_data = locate("*.fastq", illumina_dir)
self.assertTrue(not test_data == [])
sanger_dir = tempfile.mkdtemp()
out_files = [groom(x, in_qual="fastq-illumina", out_dir=sanger_dir) for
x in test_data]
self.assertTrue(all(map(file_exists, out_files)))
