def blast_db(self):
"""
Create the BLAST database of the genomes
"""
logging.info('Creating BLAST databases from sequence files')
for query_file in self.query_files:
# Remove the file extension from the file name
output = os.path.splitext(query_file)[0]
cmd = 'makeblastdb -in {fasta} -parse_seqids -max_file_sz 2GB -dbtype nucl -out {output}' \
.format(fasta=query_file,
output=output)
# Check if database already exists
if not os.path.isfile('{output}.nhr'.format(output=output)):
run_subprocess(cmd)
def primer_trim(self):
"""
Use cutadapt to trim the raw FASTQ sequence with the 5` and 3` primer files
"""
# Run cutadapt with the following arguments: -g: 5` adapters, -a: 3` adapters, -n: remove up to two adapters
# from the read,
for sample in self.samples:
# Create the necessary attributes
sample.trimmed_fastq = os.path.join(
self.sequencepath, '{sn}_trimmed.fastq'.format(sn=sample.name))
sample.cutadapt_report = os.path.join(
self.sequencepath,
'{sn}_cutadapt_report.txt'.format(sn=sample.name))
sample.cutadapt_out = str()
sample.cutadapt_err = str()
sample.cutadapt_cmd = 'cutadapt -g file:{forward_primers} -a file:{reverse_primers} -n 2 ' \
'{fastq_file} -o {trimmed_fastq} 1> {report}'\
.format(forward_primers=self.forward_primers,
reverse_primers=self.reverse_primers,
fastq_file=sample.fastq,
trimmed_fastq=sample.trimmed_fastq,
report=sample.cutadapt_report)
# Run the system call if the trimmed output file doesn't exist
if not os.path.isfile(sample.trimmed_fastq):
# Store the stdout and stderr to attributes
sample.cutadapt_out, sample.cutadapt_err = run_subprocess(
command=sample.cutadapt_cmd)
def sketch(self):
while True:
sample = self.sketchqueue.get()
if not os.path.isfile(sample.paratyper.sketchfile):
# Run the command
out, err = run_subprocess(sample.commands.sketch)
write_to_logfile(out='{cmd}\n{out}'.format(
cmd=sample.commands.sketch, out=out),
err=err,
logfile=self.logfile,
samplelog=sample.general.logout,
sampleerr=sample.general.logerr,
analysislog=None,
analysiserr=None)
self.sketchqueue.task_done()
def normalise_reads(self):
"""
"""
for sample in self.metadata:
sample.general.normalised_reads = os.path.join(
sample.general.outputdirectory,
'{sn}_normalised.fastq.gz'.format(sn=sample.name))
sample.commands.normalise = 'bbnorm.sh in1={forward} in2={reverse} maxdepth=50 mindepth=10 ecc=t out={out}'\
.format(forward=sample.general.fastqfiles[0],
reverse=sample.general.fastqfiles[1],
out=sample.general.normalised_reads)
if not os.path.isfile(sample.general.normalised_reads):
out, err = run_subprocess(sample.commands.normalise)
write_to_logfile(out='{cmd}\n{out}'.format(
cmd=sample.commands.normalise, out=out),
err=err,
logfile=self.logfile,
samplelog=sample.general.logout,
sampleerr=sample.general.logerr,
analysislog=None,
analysiserr=None)