def task_gc_gather(self):
dos = self.inputs
ds_out_fn = fn(self.ds_out)
fastq_out_fn = fn(self.fastq_out)
fasta_ds_fofn_fn = 'fasta.contigset.fofn'
dset_fns = [fn(v) for k, v in dos.items() if k.startswith('contigset_')]
open(fasta_ds_fofn_fn, 'w').write('\n'.join(dset_fns) + '\n')
fastq_fofn_fn = 'fastq.fofn'
dset_fns = [fn(v) for k, v in dos.items() if k.startswith('fastq_')]
open(fastq_fofn_fn, 'w').write('\n'.join(dset_fns) + '\n')
bash = r"""
python -m falcon_polish.mains.run_gc_gather \
{fasta_ds_fofn_fn} \
{fastq_fofn_fn} \
{ds_out_fn} \
{fastq_out_fn}
""".format(**locals())
bash_fn = 'run_gc_gather.sh'
#open(bash_fn, 'w').write(bash)
job_done = bash_fn + '.done'
hgap_config = self.parameters.get(
'hgap') # for now, all tasks are tmpdir, or not
write_script(bash, bash_fn, job_done)
self.generated_script_fn = bash_fn
def task_pbalign_gather(self):
o_unmapped_fn = fn(self.o_unmapped)
o_ds_fn = fn(self.o_ds)
dos = self.inputs
unmapped_fns = [fn(v) for k, v in dos.items() if k.startswith('unmapped')]
dset_fns = [fn(v) for k, v in dos.items() if k.startswith('alignmentset')]
unmapped_fofn_fn = 'unmapped.fofn'
open(unmapped_fofn_fn, 'w').write('\n'.join(unmapped_fns) + '\n')
ds_fofn_fn = 'gathered.alignmentsets.fofn'
open(ds_fofn_fn, 'w').write('\n'.join(dset_fns) + '\n')
bash = r"""
#rm -f {o_ds_fn}
python -m falcon_polish.mains.run_pbalign_gather \
{unmapped_fofn_fn} \
{o_unmapped_fn} \
{ds_fofn_fn} \
{o_ds_fn}
#pbvalidate {o_ds_fn}
""".format(**locals())
bash_fn = 'run_pbalign_gather.sh'
#open(bash_fn, 'w').write(bash)
job_done = bash_fn + '.done'
hgap_config = self.parameters.get(
'hgap') # for now, all tasks are tmpdir, or not
write_script(bash, bash_fn, job_done)
self.generated_script_fn = bash_fn
def task_genomic_consensus(self):
alignmentset_fn = fn(self.alignmentset)
referenceset_fn = fn(self.referenceset)
polished_fastq_fn = fn(self.polished_fastq)
variants_gff_fn = fn(self.variants_gff)
consensus_contigset_fn = fn(self.consensus_contigset)
task_opts = self.parameters['variantCaller']
options = task_opts.get('options', '')
if '--alignmentSetRefWindows' not in options:
options += ' --alignmentSetRefWindows'
fasta_fn = re.sub(".contigset.xml", ".fasta", consensus_contigset_fn)
# Possibly we should escape '{options}'
bash = """
python -m falcon_polish.mains.run_variantCaller --log-level DEBUG --options '{options}' \
{alignmentset_fn} \
{referenceset_fn} \
{polished_fastq_fn} \
{variants_gff_fn} \
{fasta_fn} \
{consensus_contigset_fn}
""".format(**locals())
bash_fn = 'run_gc.sh'
#open(bash_fn, 'w').write(bash)
job_done = bash_fn + '.done'
hgap_config = self.parameters.get('hgap') # for now, all tasks are tmpdir, or not
write_script(bash, bash_fn, job_done)
self.generated_script_fn = bash_fn
def task_polished_assembly_report(self):
task_opts = self.parameters['pbreports.tasks.summarize_coverage']
options = task_opts.get('options', '')
referenceset_fn = fn(self.referenceset)
gathered_alignmentset_fn = fn(self.gathered_alignmentset)
polished_fastq_fn = fn(self.polished_fastq)
report_fn = fn(self.report_json)
alignment_summary_gff_fn = 'alignment.summary.gff'
"""
If necessary, we could call this:
from pbreports.report.summarize_coverage.summarize_coverage import summarize_coverage
summarize_coverage(args.aln_set, args.aln_summ_gff, args.ref_set,
args.num_regions, args.region_size,
args.force_num_regions)
"""
# https://github.com/PacificBiosciences/pbreports/pull/186
bash = r"""
python -m pbreports.report.summarize_coverage.summarize_coverage \
{options} \
{gathered_alignmentset_fn} \
{referenceset_fn} \
{alignment_summary_gff_fn}
python -m pbreports.report.polished_assembly \
{alignment_summary_gff_fn} \
{polished_fastq_fn} \
{report_fn}
""".format(**locals())
bash_fn = 'run_report.sh'
#open(bash_fn, 'w').write(bash)
job_done = bash_fn + '.done'
hgap_config = self.parameters.get('hgap') # for now, all tasks are tmpdir, or not
write_script(bash, bash_fn, job_done)
self.generated_script_fn = bash_fn
def task_pbalign_gather(self):
o_unmapped_fn = fn(self.o_unmapped)
o_ds_fn = fn(self.o_ds)
dos = self.inputs
unmapped_fns = [fn(v) for k,v in dos.items() if k.startswith('unmapped')]
dset_fns = [fn(v) for k,v in dos.items() if k.startswith('alignmentset')]
unmapped_fofn_fn = 'unmapped.fofn'
open(unmapped_fofn_fn, 'w').write('\n'.join(unmapped_fns) + '\n')
ds_fofn_fn = 'gathered.alignmentsets.fofn'
open(ds_fofn_fn, 'w').write('\n'.join(dset_fns) + '\n')
bash = r"""
#rm -f {o_ds_fn}
python -m falcon_polish.mains.run_pbalign_gather \
{unmapped_fofn_fn} \
{o_unmapped_fn} \
{ds_fofn_fn} \
{o_ds_fn}
#pbvalidate {o_ds_fn}
""".format(**locals())
bash_fn = 'run_pbalign_gather.sh'
#open(bash_fn, 'w').write(bash)
job_done = bash_fn + '.done'
hgap_config = self.parameters.get('hgap') # for now, all tasks are tmpdir, or not
write_script(bash, bash_fn, job_done)
self.generated_script_fn = bash_fn
def task_gc_gather(self):
dos = self.inputs
ds_out_fn = fn(self.ds_out)
fastq_out_fn = fn(self.fastq_out)
fasta_ds_fofn_fn = 'fasta.contigset.fofn'
dset_fns = [fn(v) for k,v in dos.items() if k.startswith('contigset_')]
open(fasta_ds_fofn_fn, 'w').write('\n'.join(dset_fns) + '\n')
fastq_fofn_fn = 'fastq.fofn'
dset_fns = [fn(v) for k,v in dos.items() if k.startswith('fastq_')]
open(fastq_fofn_fn, 'w').write('\n'.join(dset_fns) + '\n')
bash = r"""
python -m falcon_polish.mains.run_gc_gather \
{fasta_ds_fofn_fn} \
{fastq_fofn_fn} \
{ds_out_fn} \
{fastq_out_fn}
""".format(**locals())
bash_fn = 'run_gc_gather.sh'
#open(bash_fn, 'w').write(bash)
job_done = bash_fn + '.done'
hgap_config = self.parameters.get('hgap') # for now, all tasks are tmpdir, or not
write_script(bash, bash_fn, job_done)
self.generated_script_fn = bash_fn
def task_genomic_consensus(self):
alignmentset_fn = fn(self.alignmentset)
referenceset_fn = fn(self.referenceset)
polished_fastq_fn = fn(self.polished_fastq)
variants_gff_fn = fn(self.variants_gff)
consensus_contigset_fn = fn(self.consensus_contigset)
task_opts = self.parameters['variantCaller']
options = task_opts.get('options', '')
if '--alignmentSetRefWindows' not in options:
options += ' --alignmentSetRefWindows'
fasta_fn = re.sub(".contigset.xml", ".fasta", consensus_contigset_fn)
# Possibly we should escape '{options}'
bash = """
python -m falcon_polish.mains.run_variantCaller --log-level DEBUG --options '{options}' \
{alignmentset_fn} \
{referenceset_fn} \
{polished_fastq_fn} \
{variants_gff_fn} \
{fasta_fn} \
{consensus_contigset_fn}
""".format(**locals())
bash_fn = 'run_gc.sh'
#open(bash_fn, 'w').write(bash)
job_done = bash_fn + '.done'
hgap_config = self.parameters.get(
'hgap') # for now, all tasks are tmpdir, or not
write_script(bash, bash_fn, job_done)
self.generated_script_fn = bash_fn
def task_polished_assembly_report(self):
task_opts = self.parameters['pbreports.tasks.summarize_coverage']
options = task_opts.get('options', '')
referenceset_fn = fn(self.referenceset)
gathered_alignmentset_fn = fn(self.gathered_alignmentset)
polished_fastq_fn = fn(self.polished_fastq)
report_fn = fn(self.report_json)
alignment_summary_gff_fn = 'alignment.summary.gff'
"""
If necessary, we could call this:
from pbreports.report.summarize_coverage.summarize_coverage import summarize_coverage
summarize_coverage(args.aln_set, args.aln_summ_gff, args.ref_set,
args.num_regions, args.region_size,
args.force_num_regions)
"""
# https://github.com/PacificBiosciences/pbreports/pull/186
bash = r"""
python -m pbreports.report.summarize_coverage.summarize_coverage \
{options} \
{gathered_alignmentset_fn} \
{referenceset_fn} \
{alignment_summary_gff_fn}
python -m pbreports.report.polished_assembly \
{alignment_summary_gff_fn} \
{polished_fastq_fn} \
{report_fn}
""".format(**locals())
bash_fn = 'run_report.sh'
#open(bash_fn, 'w').write(bash)
job_done = bash_fn + '.done'
hgap_config = self.parameters.get(
'hgap') # for now, all tasks are tmpdir, or not
write_script(bash, bash_fn, job_done)
self.generated_script_fn = bash_fn
def task_fasta2referenceset(self):
i_falcon_link_done_fn = fn(self.falcon_link_done)
idir, done_fn = os.path.split(i_falcon_link_done_fn)
i_fasta_fn = os.path.join(idir, 'asm.fasta') # by convention
o_referenceset_fn = fn(self.referenceset)
bash = """
rm -f {o_referenceset_fn} {i_fasta_fn}.fai
python -m falcon_polish.mains.run_fasta2referenceset {i_fasta_fn} {o_referenceset_fn}
""".format(**locals())
bash_fn = 'run_fasta2referenceset.sh'
#open(bash_fn, 'w').write(bash)
job_done = bash_fn + '.done'
hgap_config = self.parameters.get('hgap') # for now, all tasks are tmpdir, or not
write_script(bash, bash_fn, job_done)
self.generated_script_fn = bash_fn
def task_fasta2referenceset(self):
i_falcon_link_done_fn = fn(self.falcon_link_done)
idir, done_fn = os.path.split(i_falcon_link_done_fn)
i_fasta_fn = os.path.join(idir, 'asm.fasta') # by convention
o_referenceset_fn = fn(self.referenceset)
bash = """
rm -f {o_referenceset_fn} {i_fasta_fn}.fai
python -m falcon_polish.mains.run_fasta2referenceset {i_fasta_fn} {o_referenceset_fn}
""".format(**locals())
bash_fn = 'run_fasta2referenceset.sh'
#open(bash_fn, 'w').write(bash)
job_done = bash_fn + '.done'
hgap_config = self.parameters.get(
'hgap') # for now, all tasks are tmpdir, or not
write_script(bash, bash_fn, job_done)
self.generated_script_fn = bash_fn
def task_pbalign(self):
"""pbalign will eventually call blasr, like this:
BlasrService: Align reads to references using blasr.
BlasrService: Call "blasr /pbi/dept/secondary/siv/testdata/SA3-DS/lambda/2372215/0007_tiny/Analysis_Results/m150404_101626_42267_c100807920800000001823174110291514_s1_p0.all.subreadset.xml /home/UNIXHOME/cdunn/repo/pb/smrtanalysis-client/smrtanalysis/siv/testkit-jobs/sa3_pipelines/polished-falcon-lambda-007-tiny/job_output/tasks/falcon_ns.tasks.task_falcon2_run_asm-0/file.fasta -out /scratch/tmpTbV4Ec/wLCUdL.bam -bam -bestn 10 -minMatch 12 -nproc 16 -minSubreadLength 50 -minAlnLength 50 -minPctSimilarity 70 -minPctAccuracy 70 -hitPolicy randombest -randomSeed 1 -minPctSimilarity 70.0 "
FilterService: Filter alignments using samFilter.
FilterService: Call "rm -f /scratch/tmpTbV4Ec/aM1Mor.bam && ln -s /scratch/tmpTbV4Ec/wLCUdL.bam /scratch/tmpTbV4Ec/aM1Mor.bam"
BamPostService: Sort and build index for a bam file.
BamPostService: Call "samtools sort -m 4G /scratch/tmpTbV4Ec/aM1Mor.bam /home/UNIXHOME/cdunn/repo/pb/smrtanalysis-client/smrtanalysis/siv/testkit-jobs/sa3_pipelines/polished-falcon-lambda-007-tiny/job_output/tasks/pbalign.tasks.pbalign-0/aligned.subreads.alignmentset"
BamPostService: Call "samtools index /home/UNIXHOME/cdunn/repo/pb/smrtanalysis-client/smrtanalysis/siv/testkit-jobs/sa3_pipelines/polished-falcon-lambda-007-tiny/job_output/tasks/pbalign.tasks.pbalign-0/aligned.subreads.alignmentset.bam /home/UNIXHOME/cdunn/repo/pb/smrtanalysis-client/smrtanalysis/siv/testkit-jobs/sa3_pipelines/polished-falcon-lambda-007-tiny/job_output/tasks/pbalign.tasks.pbalign-0/aligned.subreads.alignmentset.bam.bai"
BamPostService: Call "pbindex /home/UNIXHOME/cdunn/repo/pb/smrtanalysis-client/smrtanalysis/siv/testkit-jobs/sa3_pipelines/polished-falcon-lambda-007-tiny/job_output/tasks/pbalign.tasks.pbalign-0/aligned.subreads.alignmentset.bam"
] OutputService: Generating the output XML file
"""
reads_fn = fn(self.dataset)
referenceset_fn = fn(self.referenceset)
o_alignmentset_fn = fn(self.alignmentset)
o_unmapped_fn = fn(self.unmapped)
tmpdir = tempfile.tempdir
task_opts = self.parameters['pbalign']
options = task_opts.get('options', '')
algorithmOptions = task_opts.get('algorithmOptions', '')
# Write a file of absolute paths. But that file is relative to CWD:
abs_alignmentset_fn = 'abs.' + o_alignmentset_fn
#'--debug', # requires 'ipdb'
#'--profile', # kinda interesting, but maybe slow?
#'--algorithmOptions "-minMatch 12 -bestn 10 -minPctSimilarity 70.0"',
#'--concordant',
#'--hitPolicy randombest',
#'--minAccuracy 70.0',
#'--minLength 50',
bash = """
o_fn={o_alignmentset_fn}
#rm -f ${{o_fn%.*}}
pbalign --verbose --nproc 16 {options} --tmpDir {tmpdir} --algorithmOptions "{algorithmOptions}" --unaligned {o_unmapped_fn} {reads_fn} {referenceset_fn} {o_alignmentset_fn}
#pbvalidate {o_alignmentset_fn}
dataset relativize {o_alignmentset_fn}
#pbvalidate {o_alignmentset_fn}
""".format(**locals())
bash_fn = 'run_pbalign.sh'
#open(bash_fn, 'w').write(bash)
job_done = bash_fn + '.done'
hgap_config = self.parameters.get(
'hgap') # for now, all tasks are tmpdir, or not
write_script(bash, bash_fn, job_done)
self.generated_script_fn = bash_fn
def task_pbalign(self):
"""pbalign will eventually call blasr, like this:
BlasrService: Align reads to references using blasr.
BlasrService: Call "blasr /pbi/dept/secondary/siv/testdata/SA3-DS/lambda/2372215/0007_tiny/Analysis_Results/m150404_101626_42267_c100807920800000001823174110291514_s1_p0.all.subreadset.xml /home/UNIXHOME/cdunn/repo/pb/smrtanalysis-client/smrtanalysis/siv/testkit-jobs/sa3_pipelines/polished-falcon-lambda-007-tiny/job_output/tasks/falcon_ns.tasks.task_falcon2_run_asm-0/file.fasta -out /scratch/tmpTbV4Ec/wLCUdL.bam -bam -bestn 10 -minMatch 12 -nproc 16 -minSubreadLength 50 -minAlnLength 50 -minPctSimilarity 70 -minPctAccuracy 70 -hitPolicy randombest -randomSeed 1 -minPctSimilarity 70.0 "
FilterService: Filter alignments using samFilter.
FilterService: Call "rm -f /scratch/tmpTbV4Ec/aM1Mor.bam && ln -s /scratch/tmpTbV4Ec/wLCUdL.bam /scratch/tmpTbV4Ec/aM1Mor.bam"
BamPostService: Sort and build index for a bam file.
BamPostService: Call "samtools sort -m 4G /scratch/tmpTbV4Ec/aM1Mor.bam /home/UNIXHOME/cdunn/repo/pb/smrtanalysis-client/smrtanalysis/siv/testkit-jobs/sa3_pipelines/polished-falcon-lambda-007-tiny/job_output/tasks/pbalign.tasks.pbalign-0/aligned.subreads.alignmentset"
BamPostService: Call "samtools index /home/UNIXHOME/cdunn/repo/pb/smrtanalysis-client/smrtanalysis/siv/testkit-jobs/sa3_pipelines/polished-falcon-lambda-007-tiny/job_output/tasks/pbalign.tasks.pbalign-0/aligned.subreads.alignmentset.bam /home/UNIXHOME/cdunn/repo/pb/smrtanalysis-client/smrtanalysis/siv/testkit-jobs/sa3_pipelines/polished-falcon-lambda-007-tiny/job_output/tasks/pbalign.tasks.pbalign-0/aligned.subreads.alignmentset.bam.bai"
BamPostService: Call "pbindex /home/UNIXHOME/cdunn/repo/pb/smrtanalysis-client/smrtanalysis/siv/testkit-jobs/sa3_pipelines/polished-falcon-lambda-007-tiny/job_output/tasks/pbalign.tasks.pbalign-0/aligned.subreads.alignmentset.bam"
] OutputService: Generating the output XML file
"""
reads_fn = fn(self.dataset)
referenceset_fn = fn(self.referenceset)
o_alignmentset_fn = fn(self.alignmentset)
o_unmapped_fn = fn(self.unmapped)
tmpdir = tempfile.tempdir
task_opts = self.parameters['pbalign']
options = task_opts.get('options', '')
algorithmOptions = task_opts.get('algorithmOptions', '')
# Write a file of absolute paths. But that file is relative to CWD:
abs_alignmentset_fn = 'abs.' + o_alignmentset_fn
#'--debug', # requires 'ipdb'
#'--profile', # kinda interesting, but maybe slow?
#'--algorithmOptions "-minMatch 12 -bestn 10 -minPctSimilarity 70.0"',
#'--concordant',
#'--hitPolicy randombest',
#'--minAccuracy 70.0',
#'--minLength 50',
bash = """
o_fn={o_alignmentset_fn}
#rm -f ${{o_fn%.*}}
pbalign --verbose --nproc 16 {options} --tmpDir {tmpdir} --algorithmOptions "{algorithmOptions}" --unaligned {o_unmapped_fn} {reads_fn} {referenceset_fn} {o_alignmentset_fn}
#pbvalidate {o_alignmentset_fn}
dataset relativize {o_alignmentset_fn}
#pbvalidate {o_alignmentset_fn}
""".format(**locals())
bash_fn = 'run_pbalign.sh'
#open(bash_fn, 'w').write(bash)
job_done = bash_fn + '.done'
hgap_config = self.parameters.get('hgap') # for now, all tasks are tmpdir, or not
write_script(bash, bash_fn, job_done)
self.generated_script_fn = bash_fn
def task_gc_scatter(self):
alignmentset_fn = fn(self.alignmentset)
referenceset_fn = fn(self.referenceset)
chunks_fofn_fn = fn(self.out_fofn)
config = self.parameters.get('pbcoretools.task_options', {})
max_nchunks = int(config.get('scatter_alignments_reference_max_nchunks', '13'))
bash = """
python -m falcon_polish.mains.run_gc_scatter \
--max-nchunks={max_nchunks} \
{alignmentset_fn} \
{referenceset_fn} \
{chunks_fofn_fn}
""".format(**locals())
bash_fn = 'run_gc_scatter.sh'
#open(bash_fn, 'w').write(bash)
job_done = bash_fn + '.done'
hgap_config = self.parameters.get('hgap') # for now, all tasks are tmpdir, or not
write_script(bash, bash_fn, job_done)
self.generated_script_fn = bash_fn
def task_bam2fasta_dexta(self):
i_dataset_fn = fn(self.dataset)
o_fasta_done_fn = fn(self.fasta_done)
o_fasta_fn = base_from_done(o_fasta_done_fn) + '.fasta'
prefix_basename = os.path.basename(o_fasta_fn)[:-6] #sans .fasta
prefix = prefix_basename
actual = '{}.fasta'.format(prefix) # by convention in bam2fasta
#python -m falcon_polish.mains.run_bam2fasta {i_dataset_fn} {o_fasta_fn}
bash = """
time bam2fasta -u -o {prefix} {i_dataset_fn}
time dexta -v {actual}
#time mv -f {actual}.dexta {o_fasta_fn}.dexta
touch {o_fasta_done_fn}
""".format(**locals())
bash_fn = 'run_bam2fasta.sh'
#open(bash_fn, 'w').write(bash)
job_done = bash_fn + '.done'
hgap_config = self.parameters.get('hgap') # for now, all tasks are tmpdir, or not
write_script(bash, bash_fn, job_done)
self.generated_script_fn = bash_fn
def task_bam2fasta_dexta(self):
i_dataset_fn = fn(self.dataset)
o_fasta_done_fn = fn(self.fasta_done)
o_fasta_fn = base_from_done(o_fasta_done_fn) + '.fasta'
prefix_basename = os.path.basename(o_fasta_fn)[:-6] #sans .fasta
prefix = prefix_basename
actual = '{}.fasta'.format(prefix) # by convention in bam2fasta
#python -m falcon_polish.mains.run_bam2fasta {i_dataset_fn} {o_fasta_fn}
bash = """
time bam2fasta -u -o {prefix} {i_dataset_fn}
time dexta -v {actual}
#time mv -f {actual}.dexta {o_fasta_fn}.dexta
touch {o_fasta_done_fn}
""".format(**locals())
bash_fn = 'run_bam2fasta.sh'
#open(bash_fn, 'w').write(bash)
job_done = bash_fn + '.done'
hgap_config = self.parameters.get(
'hgap') # for now, all tasks are tmpdir, or not
write_script(bash, bash_fn, job_done)
self.generated_script_fn = bash_fn
def task_pbalign_scatter(self):
"""This might have problems if run in /tmp.
"""
reads_fn = fn(self.dataset)
referenceset_fn = fn(self.referenceset)
out_json_fn = fn(self.out_json)
config = self.parameters.get('pbcoretools.task_options', {})
max_nchunks = int(config.get('scatter_subread_max_nchunks', '6'))
bash = r"""
#rm -f {out_json_fn}
python -m pbcoretools.tasks.scatter_subread_reference -v --max_nchunks={max_nchunks} \
{reads_fn} \
{referenceset_fn} \
{out_json_fn}
""".format(**locals())
bash_fn = 'run_pbalign_scatter.sh'
#open(bash_fn, 'w').write(bash)
job_done = bash_fn + '.done'
hgap_config = self.parameters.get('hgap') # for now, all tasks are tmpdir, or not
write_script(bash, bash_fn, job_done)
self.generated_script_fn = bash_fn
def task_gc_scatter(self):
alignmentset_fn = fn(self.alignmentset)
referenceset_fn = fn(self.referenceset)
chunks_fofn_fn = fn(self.out_fofn)
config = self.parameters.get('pbcoretools.task_options', {})
max_nchunks = int(
config.get('scatter_alignments_reference_max_nchunks', '13'))
bash = """
python -m falcon_polish.mains.run_gc_scatter \
--max-nchunks={max_nchunks} \
{alignmentset_fn} \
{referenceset_fn} \
{chunks_fofn_fn}
""".format(**locals())
bash_fn = 'run_gc_scatter.sh'
#open(bash_fn, 'w').write(bash)
job_done = bash_fn + '.done'
hgap_config = self.parameters.get(
'hgap') # for now, all tasks are tmpdir, or not
write_script(bash, bash_fn, job_done)
self.generated_script_fn = bash_fn
def task_pbalign_scatter(self):
"""This might have problems if run in /tmp.
"""
reads_fn = fn(self.dataset)
referenceset_fn = fn(self.referenceset)
out_json_fn = fn(self.out_json)
config = self.parameters.get('pbcoretools.task_options', {})
max_nchunks = int(config.get('scatter_subread_max_nchunks', '6'))
bash = r"""
#rm -f {out_json_fn}
python -m pbcoretools.tasks.scatter_subread_reference -v --max_nchunks={max_nchunks} \
{reads_fn} \
{referenceset_fn} \
{out_json_fn}
""".format(**locals())
bash_fn = 'run_pbalign_scatter.sh'
#open(bash_fn, 'w').write(bash)
job_done = bash_fn + '.done'
hgap_config = self.parameters.get(
'hgap') # for now, all tasks are tmpdir, or not
write_script(bash, bash_fn, job_done)
self.generated_script_fn = bash_fn