def check_gff_extension(self):
if self.gff_extension != 'gff' and self.gff_extension != 'gtf':
print(time_stamp(),
"!!WARNING!! {}'s extension is not 'gff' or 'gtf'\n".format(
self.args.gff),
flush=True)
sys.exit(1)
def run_prinseq(self):
cmd = 'seq -f %03g {0} | \
xargs -P {0} \
-I % \
prinseq-lite.pl -trim_left 5 \
-trim_right 20 \
-trim_qual_window 10 \
-trim_qual_right 20 \
-min_len 75 \
-min_qual_mean 20 \
-fastq {1}/20_fastq/FaQCs_{2}/{2}.1.trimmed.fastq.split/{2}.1.trimmed.part_%.fastq \
-fastq2 {1}/20_fastq/FaQCs_{2}/{2}.1.trimmed.fastq.split/{2}.2.trimmed.part_%.fastq \
-out_good {1}/20_fastq/prinseq_{2}/{2}.part_% \
-out_bad null \
> {1}/log/prinseq_{2}.log \
2>&1'.format(self.N_threads, self.args.out,
self.index)
try:
sbp.run(cmd,
stdout=sbp.DEVNULL,
stderr=sbp.DEVNULL,
shell=True,
check=True)
except sbp.CalledProcessError:
print(time_stamp(), '!!ERROR!! {}\n'.format(cmd), flush=True)
sys.exit(1)
def alignment(self):
aln = Alignment(args)
aln_args = []
N_files = len(self.args.rna_seq) + len(self.args.whole_genome)
N_process, each_threads = get_proc_numbers(self.args.threads, N_files)
for i, fastq in enumerate(self.args.rna_seq):
fastq1 = fastq.split(',')[0]
fastq2 = fastq.split(',')[1]
index = 'RNA-seq.0{:0>3}'.format(i)
aln_arg = '\t'.join([fastq1, fastq2, index, 'RNA'])
aln_args.append(aln_arg)
print(time_stamp(),
"{}'s prefix -> {}.".format(fastq1, index),
flush=True)
print(time_stamp(),
"{}'s prefix -> {}.".format(fastq2, index),
flush=True)
for i, fastq in enumerate(self.args.whole_genome):
fastq1 = fastq.split(',')[0]
fastq2 = fastq.split(',')[1]
index = 'WGS.0{:0>3}'.format(i)
aln_arg = '\t'.join([fastq1, fastq2, index, 'Genome'])
aln_args.append(aln_arg)
print(time_stamp(),
"{}'s prefix -> {}.".format(fastq1, index),
flush=True)
print(time_stamp(),
"{}'s prefix -> {}.".format(fastq2, index),
flush=True)
for i in range(N_files):
aln_args[i] = aln_args[i] + '\t' + str(each_threads[i])
p = Pool(N_process)
p.map(aln.run, aln_args)
p.close()
print(time_stamp(),
'alignment successfully finished.',
flush=True)
def run(self):
print(time_stamp(),
'start to annotate the reference genome.',
flush=True)
self.merge_bam()
self.transciptome_assembly()
self.make_gff()
self.run_gffread()
def run(self):
print(time_stamp(), 'start to call variants.', flush=True)
chr_names = self.get_header()
p = Pool(self.args.threads)
p.map(self.mpileup, chr_names)
p.close()
self.concat()
self.mkindex()
def __init__(self, args):
self.args = args
if self.args.disable_RNAseq_trim:
print(time_stamp(), 'disable the trimming of RNA-seq.', flush=True)
if self.args.disable_WGS_trim:
print(time_stamp(), 'disable the trimming of WGS.', flush=True)
if self.args.disable_RNAseq_trim and \
self.args.disable_WGS_trim:
print(time_stamp(), 'start to align reads.', flush=True)
else:
print(time_stamp(), 'start to trim and align reads.', flush=True)
os.mkdir('{}/20_fastq'.format(self.args.out))
os.mkdir('{}/30_bam'.format(self.args.out))
self.write_json()
os.mkdir('{0}/40_bed/'.format(self.args.out))
def run(self):
print(time_stamp(), 'start to index reference fasta.', flush=True)
cmd1 = 'hisat2-build -p {0} {1} {1} \
> {2}/log/hisat2-build.log \
2>&1'.format(self.args.threads, self.args.ref, self.args.out)
cmd2 = 'samtools faidx {} \
> {}/log/samtools_faidx.log \
2>&1'.format(self.args.ref, self.args.out)
cmd1 = clean_cmd(cmd1)
cmd2 = clean_cmd(cmd2)
print(time_stamp(), 'hisat2-build...', flush=True)
try:
sbp.run(cmd1,
stdout=sbp.DEVNULL,
stderr=sbp.DEVNULL,
shell=True,
check=True)
except sbp.CalledProcessError:
call_log(self.args.out, 'hisat2-build', cmd1)
sys.exit(1)
try:
sbp.run(cmd2,
stdout=sbp.DEVNULL,
stderr=sbp.DEVNULL,
shell=True,
check=True)
except sbp.CalledProcessError:
call_log(self.args.out, 'samtools_faidx', cmd2)
sys.exit(1)
print(time_stamp(),
'indexing of the reference genome successfully finished.',
flush=True)
def check_max_threads(self, args):
max_cpu = multi.cpu_count()
print(
time_stamp(),
'maximum number of threads which you can use is up to {}.'.format(
max_cpu),
flush=True)
if max_cpu <= args.threads:
sys.stderr.write(
('!!WARNING!! You can use up to {0} threads. '
'This program will use {0} threads.\n').format(max_cpu))
sys.stderr.flush()
args.threads = max_cpu
elif args.threads < 1:
args.threads = max_cpu
return args
def remove_duplicates(self):
cmd = 'cat {0}/candidate_genes_from_*.{1} | \
cut -f 1-9 | \
sort -u > {0}/all_candidate_genes.{1}'.format(
self.args.out, self.gff_extension)
cmd = clean_cmd(cmd)
try:
sbp.run(cmd,
stdout=sbp.DEVNULL,
stderr=sbp.DEVNULL,
shell=True,
check=True)
except sbp.CalledProcessError:
print(time_stamp(), '!!ERROR!! {}\n'.format(cmd), flush=True)
sys.exit(1)
def filter_candidates(self):
cmd = 'jiji -a {0}/50_annotation/annotation.gff \
-b {0}/40_bed \
-v {0}/60_vcf/raiden.vcf.gz \
-o {0}/70_result'.format(self.args.out)
cmd = clean_cmd(cmd)
try:
sbp.run(cmd,
stdout=sbp.DEVNULL,
stderr=sbp.DEVNULL,
shell=True,
check=True)
except sbp.CalledProcessError:
print(time_stamp(),
'!!ERROR!! {}\n'.format(cmd),
flush=True)
sys.exit(1)
shutil.move('{0}/70_result/jiji_PA_bedtools.log'.format(self.args.out),
'{0}/log/'.format(self.args.out))
shutil.move('{0}/70_result/jiji_mut_bedtools.log'.format(self.args.out),
'{0}/log/'.format(self.args.out))
def gzip_prinseq(self):
cmd1 = 'pigz -p {0} \
{1}/20_fastq/prinseq_{2}/{2}_1.fastq'.format(
self.N_threads, self.args.out, self.index)
cmd2 = 'pigz -p {0} \
{1}/20_fastq/prinseq_{2}/{2}_2.fastq'.format(
self.N_threads, self.args.out, self.index)
cmd3 = 'pigz -p {0} \
{1}/20_fastq/prinseq_{2}/{2}_1_singletons.fastq'.format(
self.N_threads, self.args.out, self.index)
cmd4 = 'pigz -p {0} \
{1}/20_fastq/prinseq_{2}/{2}_2_singletons.fastq'.format(
self.N_threads, self.args.out, self.index)
cmd1 = clean_cmd(cmd1)
cmd2 = clean_cmd(cmd2)
cmd3 = clean_cmd(cmd3)
cmd4 = clean_cmd(cmd4)
try:
sbp.run(cmd1,
stdout=sbp.DEVNULL,
stderr=sbp.DEVNULL,
shell=True,
check=True)
except sbp.CalledProcessError:
print(time_stamp(), '!!ERROR!! {}\n'.format(cmd1), flush=True)
sys.exit(1)
try:
sbp.run(cmd2,
stdout=sbp.DEVNULL,
stderr=sbp.DEVNULL,
shell=True,
check=True)
except sbp.CalledProcessError:
print(time_stamp(), '!!ERROR!! {}\n'.format(cmd2), flush=True)
sys.exit(1)
try:
sbp.run(cmd3,
stdout=sbp.DEVNULL,
stderr=sbp.DEVNULL,
shell=True,
check=True)
except sbp.CalledProcessError:
print(time_stamp(), '!!ERROR!! {}\n'.format(cmd3), flush=True)
sys.exit(1)
try:
sbp.run(cmd4,
stdout=sbp.DEVNULL,
stderr=sbp.DEVNULL,
shell=True,
check=True)
except sbp.CalledProcessError:
print(time_stamp(), '!!ERROR!! {}\n'.format(cmd4), flush=True)
sys.exit(1)
def gzip_FaQCs(self):
cmd1 = 'pigz -p {0} \
{1}/20_fastq/FaQCs_{2}/{2}.1.trimmed.fastq'.format(
self.N_threads, self.args.out, self.index)
cmd2 = 'pigz -p {0} \
{1}/20_fastq/FaQCs_{2}/{2}.2.trimmed.fastq'.format(
self.N_threads, self.args.out, self.index)
cmd3 = 'pigz -p {0} \
{1}/20_fastq/FaQCs_{2}/{2}.discard.trimmed.fastq'.format(
self.N_threads, self.args.out, self.index)
cmd4 = 'pigz -p {0} \
{1}/20_fastq/FaQCs_{2}/{2}.unpaired.trimmed.fastq'.format(
self.N_threads, self.args.out, self.index)
cmd1 = clean_cmd(cmd1)
cmd2 = clean_cmd(cmd2)
cmd3 = clean_cmd(cmd3)
cmd4 = clean_cmd(cmd4)
if os.path.isfile('{0}/20_fastq/FaQCs_{1}/{1}.1.trimmed.fastq'.format(
self.args.out, self.index)):
try:
sbp.run(cmd1,
stdout=sbp.DEVNULL,
stderr=sbp.DEVNULL,
shell=True,
check=True)
except sbp.CalledProcessError:
print(time_stamp(), '!!ERROR!! {}\n'.format(cmd1), flush=True)
sys.exit(1)
if os.path.isfile('{0}/20_fastq/FaQCs_{1}/{1}.2.trimmed.fastq'.format(
self.args.out, self.index)):
try:
sbp.run(cmd2,
stdout=sbp.DEVNULL,
stderr=sbp.DEVNULL,
shell=True,
check=True)
except sbp.CalledProcessError:
print(time_stamp(), '!!ERROR!! {}\n'.format(cmd2), flush=True)
sys.exit(1)
if os.path.isfile(
'{0}/20_fastq/FaQCs_{1}/{1}.discard.trimmed.fastq'.format(
self.args.out, self.index)):
try:
sbp.run(cmd3,
stdout=sbp.DEVNULL,
stderr=sbp.DEVNULL,
shell=True,
check=True)
except sbp.CalledProcessError:
print(time_stamp(), '!!ERROR!! {}\n'.format(cmd3), flush=True)
sys.exit(1)
if os.path.isfile(
'{0}/20_fastq/FaQCs_{1}/{1}.unpaired.trimmed.fastq'.format(
self.args.out, self.index)):
try:
sbp.run(cmd4,
stdout=sbp.DEVNULL,
stderr=sbp.DEVNULL,
shell=True,
check=True)
except sbp.CalledProcessError:
print(time_stamp(), '!!ERROR!! {}\n'.format(cmd4), flush=True)
sys.exit(1)
def main():
print(time_stamp(), 'start to filter the causal genes.', flush=True)
Jiji(args).run()
print(time_stamp(),
'Filtering process successfully finished.\n',
flush=True)
def main():
print(time_stamp(), 'start to run RaIDeN.', flush=True)
RaIDeN(args).run()
print(time_stamp(), 'RaIDeN successfully finished.\n', flush=True)
def merge_fastq(self):
cmd1 = 'cat {0}/20_fastq/prinseq_{1}/{1}.part_*_1.fastq \
> {0}/20_fastq/prinseq_{1}/{1}_1.fastq'.format(
self.args.out, self.index)
cmd2 = 'cat {0}/20_fastq/prinseq_{1}/{1}.part_*_2.fastq > \
{0}/20_fastq/prinseq_{1}/{1}_2.fastq'.format(
self.args.out, self.index)
cmd3 = 'cat {0}/20_fastq/prinseq_{1}/{1}.part_*_1_singletons.fastq \
> {0}/20_fastq/prinseq_{1}/{1}_1_singletons.fastq'.format(
self.args.out, self.index)
cmd4 = 'cat {0}/20_fastq/prinseq_{1}/{1}.part_*_2_singletons.fastq \
> {0}/20_fastq/prinseq_{1}/{1}_2_singletons.fastq'.format(
self.args.out, self.index)
cmd1 = clean_cmd(cmd1)
cmd2 = clean_cmd(cmd2)
cmd3 = clean_cmd(cmd3)
cmd4 = clean_cmd(cmd4)
try:
sbp.run(cmd1,
stdout=sbp.DEVNULL,
stderr=sbp.DEVNULL,
shell=True,
check=True)
except sbp.CalledProcessError:
print(time_stamp(), '!!ERROR!! {}\n'.format(cmd1), flush=True)
sys.exit(1)
try:
sbp.run(cmd2,
stdout=sbp.DEVNULL,
stderr=sbp.DEVNULL,
shell=True,
check=True)
except sbp.CalledProcessError:
print(time_stamp(), '!!ERROR!! {}\n'.format(cmd2), flush=True)
sys.exit(1)
try:
sbp.run(cmd3,
stdout=sbp.DEVNULL,
stderr=sbp.DEVNULL,
shell=True,
check=True)
except sbp.CalledProcessError:
print(time_stamp(), '!!ERROR!! {}\n'.format(cmd3), flush=True)
sys.exit(1)
try:
sbp.run(cmd4,
stdout=sbp.DEVNULL,
stderr=sbp.DEVNULL,
shell=True,
check=True)
except sbp.CalledProcessError:
print(time_stamp(), '!!ERROR!! {}\n'.format(cmd4), flush=True)
sys.exit(1)