def rmdup(align_file, out_file):
cmd = ("samtools view -bh {align_file} | samtools sort -o -n - {tmp} | bammarkduplicates rmdup=1 O={tx_out_file}")
tmp = align_file + "_tmp"
if not os.path.exists(out_file):
with file_transaction(out_file) as tx_out_file:
do.run(cmd.format(**locals()))
return out_file
def qc(sam_file):
"""fastqc for the sam file"""
out_dir = os.path.basename(sam_file) + "_fastq"
cmd = "fastqc {sam_file} -f sam -o {out_dir}".format(**locals())
if not os.path.exists(out_dir):
os.mkdir(out_dir)
do.run(cmd)
return out_dir
def _bowtie_align(fastq_file, control_index, out_file):
cmd = ("bowtie2 -p 4 --no-unal -x {control_index} -U {fastq_file} | samtools view -Shb /dev/stdin > {tx_out_file} ")
stat_file = out_file + ".flagstat"
if not os.path.exists(out_file):
with file_transaction(out_file) as tx_out_file:
do.run(cmd.format(**locals()), "bowtie2 %s" % fastq_file)
do.run("samtools flagstat {out_file} > {stat_file}".format(**locals()), "stats control sequences")
return stat_file
def rmdup(align_file, out_file):
cmd = (
"samtools view -bh {align_file} | samtools sort -o -n - {tmp} | bammarkduplicates rmdup=1 O={tx_out_file}"
)
tmp = align_file + "_tmp"
if not os.path.exists(out_file):
with file_transaction(out_file) as tx_out_file:
do.run(cmd.format(**locals()))
return out_file
def qc(data, args):
"""fastqc for the sam file"""
sam_file = data['align']
out_dir = os.path.basename(sam_file) + "_fastq"
cmd = "fastqc {sam_file} -f sam -o {out_dir}".format(**locals())
if not os.path.exists(out_dir):
os.mkdir(out_dir)
do.run(cmd)
else:
logger.my_logger.info("%s has already been QC, skipping." % (sam_file))
return data
def qc(data, args):
"""fastqc for the sam file"""
sam_file = data['r1_path']
out_dir = os.path.basename(data["sample_id"]) + "_fastqc"
cmd = "fastqc {sam_file} -f sam -o {out_dir}".format(**locals())
if not os.path.exists(out_dir):
os.mkdir(out_dir)
do.run(cmd)
else:
logger.my_logger.info("%s has already been QC, skipping." % (sam_file))
return data
def _bowtie_align(fastq_file, control_index, out_file):
cmd = (
"bowtie2 -p 4 --no-unal -x {control_index} -U {fastq_file} | samtools view -Shb /dev/stdin > {tx_out_file} "
)
stat_file = out_file + ".flagstat"
if not os.path.exists(out_file):
with file_transaction(out_file) as tx_out_file:
do.run(cmd.format(**locals()), "bowtie2 %s" % fastq_file)
do.run("samtools flagstat {out_file} > {stat_file}".format(**locals()),
"stats control sequences")
return stat_file
def star_align(data, args, fastq_path, out_prefix, opts=""):
cores = args.cores_per_job
reference_prefix = args.aligner_index
max_best = MAX_BEST
out_file = out_prefix + "Aligned.out.sam"
if not os.path.exists(out_file):
cmd = ("STAR --genomeDir {reference_prefix} --readFilesIn {fastq_path} --readFilesCommand zcat "
"--runThreadN {cores} --outFileNamePrefix {out_prefix} "
"--outFilterMultimapNmax {max_best} "
"--outSAMattributes NH HI NM MD AS {opts} "
"--outSAMstrandField intronMotif").format(**locals())
do.run(cmd)
clean_file = clean_align(out_file, out_prefix + "cleaned.sam")
return out_file, clean_file
def star_align(fastq_path, reference_prefix, out_prefix, cores=1):
max_best = MAX_BEST
out_file = out_prefix + "Aligned.out.sam"
if file_exists(out_file):
print ("%s has already been aligned, skipping." % (fastq_path))
return out_file
cmd = ("STAR --genomeDir {reference_prefix} --readFilesIn {fastq_path} --readFilesCommand zcat "
"--runThreadN {cores} --outFileNamePrefix {out_prefix} "
"--outFilterMultimapNmax {max_best} "
"--outSAMattributes NH HI NM MD AS "
"--outSAMstrandField intronMotif").format(**locals())
do.run(cmd)
return out_file
def star_align(data, args, fastq_path, out_prefix, opts=""):
cores = args.cores_per_job
reference_prefix = args.aligner_index
max_best = MAX_BEST
out_file = out_prefix + "Aligned.sortedByCoord.out.bam"
if not os.path.exists(out_file):
cmd = (
"STAR --genomeDir {reference_prefix} --readFilesIn {fastq_path} --readFilesCommand zcat "
"--runThreadN {cores} --outFileNamePrefix {out_prefix} "
"--outFilterMultimapNmax {max_best} --outSAMtype BAM SortedByCoordinate "
"--outSAMattributes NH HI NM MD AS {opts} "
"").format(**locals())
do.run(cmd)
clean_file = clean_align(out_file, out_prefix + "cleaned.sam")
return out_file, clean_file
def _cmd_counts(in_file, out_file, gtf_file, cores):
if not os.path.exists(out_file):
cmd = "featureCounts -R -T {cores} -a {gtf_file} -o {out_file} {in_file}"
do.run(cmd.format(**locals()))
return in_file + ".featureCounts"
def _cmd_counts(in_file, out_file, gtf_file, cores):
if not os.path.exists(out_file):
cmd = "featureCounts -R -T {cores} --primary -a {gtf_file} -o {out_file} {in_file}"
do.run(cmd.format(**locals()))
return in_file + ".featureCounts"