def __init__(self, orientation, bed, job_name, out_sh, submit, directory):
cmd_list = []
for filename in glob('{}/*sorted.bam'.format(directory)):
cmd_list.append(
'count_tags.py --annotation_file {} -f {} -b {} -o {}.count'.format(
bed, orientation, filename, filename))
sub = Submitter(queue_type='PBS', sh_filename=out_sh,
commands=cmd_list,
job_name=job_name, nodes=1, ppn=16, queue='home',
walltime='1:00:00',
array=True, max_running=20)
sub.write_sh(submit=submit)
def __init__(self, job_name, out_sh=None, queue_type='PBS',
directory='./', submit=True):
cmd_list = []
for file in glob('{}/*sam'.format(directory)):
cmd_list.append(
'samtools view -bS -q 10 {} > {}.bam'.format(file, file))
sub = Submitter(queue_type=queue_type,
sh_filename=out_sh,
commands=cmd_list,
job_name=job_name, nodes=1, ppn=1, queue='home',
walltime='1:00:00',
array=True, max_running=20)
sub.job(submit=submit)
def __init__(self, gtf, job_name, out_sh, submit, directory):
commands = []
for bam in iglob('{}/*.sorted.bam'.format(directory.rstrip('/'))):
commands.append('cufflinks --GTF {0} --GTF-guide '
'--multi-read-correct --num-threads 8 {1}'.format(
gtf, bam
))
sub = Submitter(queue_type='PBS', sh_filename=out_sh,
commands=commands,
job_name=job_name, nodes=1, ppn=8, walltime='0:30:00',
array=True,
max_running=20
)
sub.write_sh(submit=submit)
def __init__(self, job_name, out_sh=None,
directory='./', submit=True):
command_list = []
for file in glob('{}/*sorted.bam'.format(directory)):
command_list.append('samtools index {0}'.format(file))
# def submit_and_write(name, command_list):
sub = Submitter(queue_type='PBS', sh_filename=out_sh,
commands=command_list,
job_name=job_name, nodes=1, ppn=1, queue='home',
array=True,
max_running=20, walltime='0:30:00')
sub.job(submit=submit)
def __init__(self, job_name, out_sh=None,
directory='./', submit=True):
cmd_list = []
for file in glob('{}/*bam'.format(directory.rstrip('/'))):
cmd_list.append(
"samtools view -h -F 4 {0} | awk '$6 !~ /N/ || $1 ~ /@/' "
"| "
"samtools view -bS - > {0}.unspliced.bam".format(file))
sub = Submitter(queue_type='PBS', sh_filename=out_sh,
commands=cmd_list,
job_name=job_name, nodes=1, ppn=16, queue='home',
array=True,
walltime='0:30:00',
max_running=10)
sub.job(submit=submit)
def __init__(self, job_name, out_sh, directory, submit):
command_list = []
for filename in glob('{}/*bam'.format(directory)):
command_list.append(
'samtools sort -@ 8 -m 50000000000 {0} {0}.sorted'
.format(filename))
# def submit_and_write(name, command_list):
sub = Submitter(queue_type='PBS', sh_filename=out_sh,
commands=command_list,
job_name=job_name, nodes=1, ppn=8, queue='home',
array=True,
max_running=10, walltime='0:30:00')
sub.write_sh(submit=submit)
def __init__(self, job_name, out_sh, submit, directory):
cmd_list = []
for file in glob('{}/*count'.format(directory.rstrip('/'))):
cmd_list.append('single_RPKM.py -i {} -o {}.rpkm'.format(
file, file))
sub = Submitter(queue_type='PBS',
sh_filename=out_sh,
commands=cmd_list,
job_name=job_name,
nodes=1,
ppn=1,
queue='home',
walltime='1:00:00',
array=True,
max_running=20)
sub.write_sh(submit=submit)
def __init__(self, orientation, bed, job_name, out_sh, submit, directory):
cmd_list = []
for filename in glob('{}/*sorted.bam'.format(directory)):
cmd_list.append(
'count_tags.py --annotation_file {} -f {} -b {} -o {}.count'.
format(bed, orientation, filename, filename))
sub = Submitter(queue_type='PBS',
sh_filename=out_sh,
commands=cmd_list,
job_name=job_name,
nodes=1,
ppn=16,
queue='home',
walltime='1:00:00',
array=True,
max_running=20)
sub.write_sh(submit=submit)
def __init__(self, job_name, out_sh=None, directory='./', submit=True):
cmd_list = []
for file in glob('{}/*bam'.format(directory.rstrip('/'))):
cmd_list.append(
"samtools view -h -F 4 {0} | awk '$6 !~ /N/ || $1 ~ /@/' "
"| "
"samtools view -bS - > {0}.unspliced.bam".format(file))
sub = Submitter(queue_type='PBS',
sh_filename=out_sh,
commands=cmd_list,
job_name=job_name,
nodes=1,
ppn=16,
queue='home',
array=True,
walltime='0:30:00',
max_running=10)
sub.job(submit=submit)
def __init__(self, job_name, out_sh=None, directory='./', submit=True):
cmd_list = []
for filename in glob('{}/*fastq'.format(directory)):
cmd_list.append('bowtie \
-c \
-S \
-q \
-p 16 \
-e 100 \
-l 20 \
--un {0}.norep \
all_ref \
{0} \
| grep -v \"@\" \
| perl /home/ppliu/tscc_scripts/count_aligned_from_sam.pl \
> {0}.repeat_counts'.format(filename))
for filename in glob('{}/*gz'.format(directory)):
cmd_list.append('gunzip -c {0} \
|bowtie \
-c \
-S \
-q \
-p 16 \
-e 100 \
-l 20 \
--un {0}.norep \
all_ref \
- \
| grep -v \"@\" \
| perl /home/ppliu/tscc_scripts/count_aligned_from_sam.pl \
> {0}.repeat_counts'.format(filename))
sub = Submitter(queue_type='PBS',
sh_filename=out_sh,
commands=cmd_list,
job_name=job_name,
nodes=1,
ppn=16,
walltime='2:30:00',
array=True,
max_running=20)
sub.write_sh(submit=submit)
def __init__(self, job_name, out_sh=None, directory='./', submit=True):
command_list = []
for file in glob('{}/*sorted.bam'.format(directory)):
command_list.append('samtools index {0}'.format(file))
# def submit_and_write(name, command_list):
sub = Submitter(queue_type='PBS',
sh_filename=out_sh,
commands=command_list,
job_name=job_name,
nodes=1,
ppn=1,
queue='home',
array=True,
max_running=20,
walltime='0:30:00')
sub.job(submit=submit)
def __init__(self, job_name, out_sh=None,
directory='./', submit=True):
cmd_list = []
for filename in glob('{}/*fastq'.format(directory)):
cmd_list.append('bowtie \
-c \
-S \
-q \
-p 16 \
-e 100 \
-l 20 \
--un {0}.norep \
all_ref \
{0} \
| grep -v \"@\" \
| perl /home/ppliu/tscc_scripts/count_aligned_from_sam.pl \
> {0}.repeat_counts'.format(filename))
for filename in glob('{}/*gz'.format(directory)):
cmd_list.append('gunzip -c {0} \
|bowtie \
-c \
-S \
-q \
-p 16 \
-e 100 \
-l 20 \
--un {0}.norep \
all_ref \
- \
| grep -v \"@\" \
| perl /home/ppliu/tscc_scripts/count_aligned_from_sam.pl \
> {0}.repeat_counts'.format(filename))
sub = Submitter(queue_type='PBS', sh_filename=out_sh,
commands=cmd_list, job_name=job_name, nodes=1,
ppn=16, walltime='2:30:00',
array=True,
max_running=20
)
sub.write_sh(submit=submit)
def __init__(self, job_name, out_sh, submit=False, directory='./'):
try:
os.mkdir('{}/filtered/'.format(directory.rstrip('/')))
except OSError:
pass
commands = []
for filename in iglob('{}/*.fastq.gz'.format(directory.rstrip('/'))):
#TODO: the -l argument "20" should be a % of read length
commands.append('echo {0}; zcat {0} | fastx_artifacts_filter | '
'fastq_quality_trimmer -l 20 -t 30 | '
'fastq_quality_filter -q 30 -p 90 -z '
'> filtered/{0}'.format(filename))
sub = Submitter(queue_type='PBS', sh_filename=out_sh,
commands=commands,
job_name=job_name, nodes=1, ppn=2, queue='home',
array=True,
max_running=20, walltime='1:00:00')
sub.write_sh(submit=submit)
def __init__(self,
job_name,
out_sh=None,
queue_type='PBS',
directory='./',
submit=True):
cmd_list = []
for file in glob('{}/*sam'.format(directory)):
cmd_list.append('samtools view -bS -q 10 {} > {}.bam'.format(
file, file))
sub = Submitter(queue_type=queue_type,
sh_filename=out_sh,
commands=cmd_list,
job_name=job_name,
nodes=1,
ppn=1,
queue='home',
walltime='1:00:00',
array=True,
max_running=20)
sub.job(submit=submit)
def __init__(self, genome, out_dir='./', directory='./', submit=True,
ppn=8, job_name='STAR', out_sh='STAR.sh', walltime='0:30:00',
outReadsUnmapped='Fastx', outFilterMismatchNmax=5,
outFilterMismatchNoverLmax=0.3, outFilterMultimapNmax=5,
outFilterScoreMin=10, outFilterType='BySJout',
outSAMattributes='All',
outSAMstrandField='intronMotif',
clip5pNbases=0, clip3pNbases=0, additional_STAR_args='', extension='.gz'):
"""Read the fastq files in a directory, assuming that the first 2
underscore-separated parts of a filename are the unique sample ID,
then running STAR. Most of these arguments are the defaults in STAR,
except:
outReadsUnmapped : str
'Fastx' instead of 'None' so the unmapped reads can be remapped
to the spikein genomes, for example
outFilterMismatchNmax : int
5 instead of 10
outFilterMultimapNmax : int
5 instead of 10
outFilterType : str
'BySJout' instead of 'None', so that all junction reads pass our
stringent filter of at least 4bp overhang for annotated and at
least 8bp overhang for unannotated
outSAMattributes : str
'All' instead of 'None' for more information just in case
outSAMstrandField : str
'intronMotif' instead of 'None' for compatibility with Cufflinks
"""
commands = []
# Make the directory
try:
os.mkdir(out_dir)
except OSError:
# It's already there, don't do anything
pass
# Set of unique sample ids for checking if we've read them all
sample_ids = set([])
for read1 in iglob('{}/*R1*{}'.format(directory.rstrip('/'), extension)):
# if read1.endswith('gz'):
# compressed = True
# else:
# compressed = False
# readFilesCommand = 'zcat' if compressed else 'cat'
# Remove trailing "A" and "B" so they get merged
sample_id = '_'.join(os.path.basename(read1).split('.')[0].split(
'_')[:2]).rstrip(
'ABCDEFGH')
if sample_id in sample_ids:
continue
paired = os.path.isfile(read1.replace('R1', 'R2'))
print sample_id, 'paired', paired
read1 = ','.join(glob('{}*R1*{}'.format(sample_id, extension)))
read2 = read1.replace('R1', 'R2') if paired else ""
print 'R1', read1
print 'R2', read2
sample_ids.add(sample_id)
# print sample_id
commands.append('''STAR \
--runMode alignReads \
--runThreadN {0} \
--genomeDir {1} \
--genomeLoad LoadAndRemove \
--readFilesCommand zcat \
--readFilesIn {2} {3} \
--outFileNamePrefix {4}/{5}. \
--outReadsUnmapped {6} \
--outFilterMismatchNmax {7} \
--outFilterMismatchNoverLmax {8} \
--outFilterMultimapNmax {9} \
--outFilterScoreMin {10} \
--outFilterType {11} \
--outSAMattributes {12} \
--outSAMstrandField {13} \
--clip5pNbases {14} \
--clip3pNbases {15} \
{16}'''.format(ppn,
genome,
read1,
read2,
out_dir.rstrip('/'),
sample_id,
outReadsUnmapped,
outFilterMismatchNmax,
outFilterMismatchNoverLmax,
outFilterMultimapNmax,
outFilterScoreMin,
outFilterType,
outSAMattributes,
outSAMstrandField,
clip5pNbases,
clip3pNbases,
additional_STAR_args))
sub = Submitter(queue_type='PBS', sh_filename=out_sh,
commands=commands,
job_name=job_name, nodes=1, ppn=ppn,
array=True, max_running=20,
queue='home', walltime=walltime)
sub.write_sh(submit=submit)
def __init__(self,
genome,
out_dir='./',
directory='./',
submit=True,
ppn=8,
job_name='STAR',
out_sh='STAR.sh',
walltime='0:30:00',
outReadsUnmapped='Fastx',
outFilterMismatchNmax=5,
outFilterMismatchNoverLmax=0.3,
outFilterMultimapNmax=5,
outFilterScoreMin=10,
outFilterType='BySJout',
outSAMattributes='All',
outSAMstrandField='intronMotif',
clip5pNbases=0,
clip3pNbases=0,
additional_STAR_args='',
extension='.gz'):
"""Read the fastq files in a directory, assuming that the first 2
underscore-separated parts of a filename are the unique sample ID,
then running STAR. Most of these arguments are the defaults in STAR,
except:
outReadsUnmapped : str
'Fastx' instead of 'None' so the unmapped reads can be remapped
to the spikein genomes, for example
outFilterMismatchNmax : int
5 instead of 10
outFilterMultimapNmax : int
5 instead of 10
outFilterType : str
'BySJout' instead of 'None', so that all junction reads pass our
stringent filter of at least 4bp overhang for annotated and at
least 8bp overhang for unannotated
outSAMattributes : str
'All' instead of 'None' for more information just in case
outSAMstrandField : str
'intronMotif' instead of 'None' for compatibility with Cufflinks
"""
commands = []
# Make the directory
try:
os.mkdir(out_dir)
except OSError:
# It's already there, don't do anything
pass
# Set of unique sample ids for checking if we've read them all
sample_ids = set([])
for read1 in iglob('{}/*R1*{}'.format(directory.rstrip('/'),
extension)):
# if read1.endswith('gz'):
# compressed = True
# else:
# compressed = False
# readFilesCommand = 'zcat' if compressed else 'cat'
# Remove trailing "A" and "B" so they get merged
sample_id = '_'.join(
os.path.basename(read1).split('.')[0].split('_')[:2]).rstrip(
'ABCDEFGH')
if sample_id in sample_ids:
continue
paired = os.path.isfile(read1.replace('R1', 'R2'))
print sample_id, 'paired', paired
read1 = ','.join(glob('{}*R1*{}'.format(sample_id, extension)))
read2 = read1.replace('R1', 'R2') if paired else ""
print 'R1', read1
print 'R2', read2
sample_ids.add(sample_id)
# print sample_id
commands.append('''STAR \
--runMode alignReads \
--runThreadN {0} \
--genomeDir {1} \
--genomeLoad LoadAndRemove \
--readFilesCommand zcat \
--readFilesIn {2} {3} \
--outFileNamePrefix {4}/{5}. \
--outReadsUnmapped {6} \
--outFilterMismatchNmax {7} \
--outFilterMismatchNoverLmax {8} \
--outFilterMultimapNmax {9} \
--outFilterScoreMin {10} \
--outFilterType {11} \
--outSAMattributes {12} \
--outSAMstrandField {13} \
--clip5pNbases {14} \
--clip3pNbases {15} \
{16}'''.format(ppn, genome, read1, read2, out_dir.rstrip('/'),
sample_id, outReadsUnmapped, outFilterMismatchNmax,
outFilterMismatchNoverLmax, outFilterMultimapNmax,
outFilterScoreMin, outFilterType, outSAMattributes,
outSAMstrandField, clip5pNbases, clip3pNbases,
additional_STAR_args))
sub = Submitter(queue_type='PBS',
sh_filename=out_sh,
commands=commands,
job_name=job_name,
nodes=1,
ppn=ppn,
array=True,
max_running=20,
queue='home',
walltime=walltime)
sub.write_sh(submit=submit)
