def run(self):
cmd = [
'$MANTA/bin/configManta.py', '--exome', '--referenceFasta',
self.cfg['fasta_file'], '--normalBam',
self.input()['N']['preprocess']['bam'].path, '--tumorBam',
self.input()['T']['preprocess']['bam'].path, '--rundir',
os.path.join(self.cfg['output_dir'], self.case, 'variant_prep',
'manta')
] # TODO add logic for exome to handle multiple sequencing preps
# pipeline_utils.cluster_command_call(self, cmd, threads=1, ram=1, cfg=self.cfg)
pipeline_utils.command_call(cmd)
cmd = [
os.path.join(self.cfg['output_dir'], self.case, 'variant_prep',
'manta', 'runWorkflow.py'), '-j', self.max_threads
]
if self.cfg['cluster_exec']:
pipeline_utils.cluster_command_call(
self,
cmd,
threads=self.cfg['max_threads'],
ram=12,
cfg=self.cfg,
err_log=self.output()['err_log'].path)
else:
pipeline_utils.command_call(cmd,
err_log=self.output()['err_log'].path)
def run(self):
if self.input()['filter_germline']['filter_germline'].path.endswith(
'.gz'):
input_vcf = self.input(
)['filter_germline']['filter_germline'].path.split('.gz')[0]
with gzip.open(
self.input()['filter_germline']['filter_germline'].path,
'rb') as vcf_in, open(input_vcf, 'wb') as vcf_out:
shutil.copyfileobj(vcf_in, vcf_out)
else:
input_vcf = self.input(
)['filter_germline']['filter_germline'].path.endswith('.gz')
cmd = [
'perl', '/root/pipeline/code/source/vcf2maf/vcf2maf.pl',
'--ref-fasta', self.cfg['fasta_file'], '--vep-forks',
self.cfg['max_threads'], '--input-vcf', input_vcf, '--output-maf',
self.output()['vcf2maf'].path, '--tumor-id',
'%s_N' % self.case
]
if self.cfg['cluster_exec']:
pipeline_utils.cluster_command_call(
self,
cmd,
threads=self.cfg['max_threads'],
ram=16,
cfg=self.cfg,
err_log=self.output()['err_log'].path)
else:
pipeline_utils.command_call(cmd,
err_log=self.output()['err_log'].path)
def run(self):
if self.input()['filter_mutect2']['filter_mutect2'].path.endswith(
'.gz'):
input_vcf = self.input(
)['filter_mutect2']['filter_mutect2'].path.split('.gz')[0]
with gzip.open(
self.input()['filter_mutect2']['filter_mutect2'].path,
'rb') as vcf_in, open(input_vcf, 'wb') as vcf_out:
shutil.copyfileobj(vcf_in, vcf_out)
else:
input_vcf = self.input(
)['filter_mutect2']['filter_mutect2'].path.endswith('.gz')
cmd = [
'fpfilter', '--vcf-file', input_vcf, '--bam-file',
self.input()['preprocess']['bam'].path, '--reference',
self.cfg['fasta_file'], '--sample', self.case + '_T', '--output',
self.output()['fpfilter'].path
]
if self.cfg['cluster_exec']:
pipeline_utils.cluster_command_call(
self,
cmd,
threads=self.cfg['max_threads'],
ram=16,
cfg=self.cfg,
err_log=self.output()['err_log'].path)
else:
pipeline_utils.command_call(cmd,
err_log=self.output()['err_log'].path)
def run(self):
cmd = ['java', '-jar', '$PICARD', 'CreateSequenceDictionary', 'R=%s' % self.cfg['fasta_file'], 'O=%s' % self.output().path]
if self.cfg['cluster_exec']:
pipeline_utils.cluster_command_call(self, cmd, threads=1, ram=5, cfg=self.cfg) #, err_log=self.output()['err_log'].path)
else:
pipeline_utils.command_call(cmd) #, err_log=self.output()['err_log'].path)
shutil.copyfile(self.output()['picard_index'].path, self.cfg['fasta_file'].split('.fa')[0] + '.dict')
def run(self):
cmd = [
'java', '-Dsamjdk.use_async_io_read_samtools=false',
'-Dsamjdk.use_async_io_write_samtools=true',
'-Dsamjdk.use_async_io_write_tribble=false',
'-Dsamjdk.compression_level=2',
'-Djava.io.tmpdir=%s' % self.cfg['tmp_dir'], '-jar',
'/root/pipeline/resources/broad/gatk-4.0.12.0/gatk-package-4.0.12.0-local.jar',
'CreateSomaticPanelOfNormals', '-O',
self.output()['mutect2_pon'].path, '--tmp-dir', self.cfg['tmp_dir']
]
for case in self.input():
cmd += [
'-vcfs',
self.input()[case]['mutect2_normal']['mutect2_normal'].path
]
if self.cfg['cluster_exec']:
pipeline_utils.cluster_command_call(
self,
cmd,
threads=1,
ram=12,
cfg=self.cfg,
err_log=self.output()['err_log'].path)
else:
pipeline_utils.command_call(cmd,
err_log=self.output()['err_log'].path)
def run(self):
cmd = [
'$STRELKA/bin/configureStrelkaSomaticWorkflow.py', '--exome',
'--referenceFasta', self.cfg['fasta_file'], '--normalBam',
self.input()['N']['preprocess']['bam'].path, '--tumorBam',
self.input()['T']['preprocess']['bam'].path, '--indelCandidates',
self.input()['manta']['manta'].path, '--rundir',
os.path.join(self.cfg['output_dir'], self.case, 'variant_prep',
'strelka')
]
# pipeline_utils.cluster_command_call(self, cmd, threads=1, ram=1, cfg=self.cfg)
pipeline_utils.command_call(cmd)
cmd = [
os.path.join(self.cfg['output_dir'], self.case, 'variant_prep',
'strelka', 'runWorkflow.py'), '-m', 'local', '-j',
self.cfg['max_threads']
]
if self.cfg['cluster_exec']:
pipeline_utils.cluster_command_call(
self,
cmd,
threads=self.cfg['max_threads'],
ram=12,
cfg=self.cfg,
err_log=self.output()['err_log'].path)
else:
pipeline_utils.command_call(cmd,
err_log=self.output()['err_log'].path)
def run(self): cmd = ['trim_galore', '--fastqc', '--fastqc_args "--outdir %s"' % os.path.dirname(self.output()['fastqc'][0].path), '--paired', '-o', os.path.dirname(self.output()['trimgalore'][0].path), '--basename', '%s_%s_%s' % (self.case, self.sample, self.lane), '--gzip', self.cfg['cases'][self.case][self.sample][self.lane]['fastq1'], self.cfg['cases'][self.case][self.sample][self.lane]['fastq2']] pipeline_utils.confirm_path(self.output()['trimgalore'][0].path) pipeline_utils.confirm_path(self.output()['fastqc'][0].path) if self.cfg['cluster_exec']: pipeline_utils.cluster_command_call(self, cmd, threads=1, ram=4, cfg=self.cfg, err_log=self.output()['err_log'].path) else: pipeline_utils.command_call(cmd, err_log=self.output()['err_log'].path)
def run(self): read_group = pipeline_utils.assign_rg(self.input()['trim']['trimgalore'][0].path, self.input()['trim']['trimgalore'][1].path, self.case, self.sample, self.cfg) if self.cfg['cluster_exec']: cmd = ['bwa', 'mem', '-M', '-t', self.cfg['max_threads'], '-R', "'%s'" % read_group, self.cfg['fasta_file'], self.input()['trim']['trimgalore'][0].path, self.input()['trim']['trimgalore'][1].path, '|', 'samtools', 'view', '-bh', '|', 'samtools', 'sort', '-o', self.output()['bwa_mem'].path] pipeline_utils.cluster_command_call(self, cmd, threads=self.cfg['max_threads'], ram=8, cfg=self.cfg, err_log=self.output()['err_log'].path) else: # cmds = [['bwa', 'mem', '-M', '-t', self.cfg['max_threads'], '-R', "'%s'" % read_group, self.cfg['fasta_file'], self.input()['trim']['trimgalore'][0].path, self.input()['trim']['trimgalore'][1].path], ['samtools', 'view', '-bh', ], ['samtools', 'sort', '-o', self.output()['bwa_mem'].path]] cmd = ['bwa', 'mem', '-M', '-t', self.cfg['max_threads'], '-R', "'%s'" % read_group, self.cfg['fasta_file'], self.input()['trim']['trimgalore'][0].path, self.input()['trim']['trimgalore'][1].path, '|', 'samtools', 'view', '-bh', '|', 'samtools', 'sort', '-o', self.output()['bwa_mem'].path] # pipeline_utils.piped_command_call(cmds, err_log=self.output()['err_log'].path) pipeline_utils.command_call(cmd, err_log=self.output()['err_log'].path)
def run(self):
if len(self.input()) > 1:
cmd = ['java', '-Djava.io.tmpdir=%s' % self.cfg['tmp_dir'], '-jar', '$PICARD', 'MergeSamFiles', 'O=%s' % self.output()['merge_bams'].path]
for lane in self.input():
cmd += ['I=%s' % self.input()[lane]['align']['bwa_mem'].path]
if self.cfg['cluster_exec']:
pipeline_utils.cluster_command_call(self, cmd, threads=1, ram=5, cfg=self.cfg, err_log=self.output()['err_log'].path)
else:
pipeline_utils.command_call(cmd, err_log=self.output()['err_log'].path)
else:
for lane in self.input():
shutil.copyfile(self.input()[lane]['align']['bwa_mem'].path, self.output()['merge_bams'].path)
with open(self.output()['err_log'].path, 'w') as f:
f.write('')
def run(self):
cmd = [
'msisensor', 'msi', '-d',
'/root/pipeline/resources/misc/hg19_microsatellites.list', '-t',
self.input()['T']['preprocess']['bam'].path, '-o',
self.output()['msisensor'].path, '-l', '1', '-q', '1', '-c', '20'
] # , '-e', self.cfg['library_bed'], '-b', self.cfg['max_threads']
if self.cfg['cluster_exec']:
pipeline_utils.cluster_command_call(
self,
cmd,
threads=self.cfg['max_threads'],
ram=16,
cfg=self.cfg,
err_log=self.output()['err_log'].path)
else:
pipeline_utils.command_call(cmd,
err_log=self.output()['err_log'].path)
def run(self):
cmd = [
'snp-pileup', '-g', '-q15', '-Q20', '-P100', '-r25,0',
self.cfg['germline_all'],
self.output()['facets_snp_pileup'].path,
self.input()['N']['preprocess']['bam'].path,
self.input()['T']['preprocess']['bam'].path
]
if self.cfg['cluster_exec']:
pipeline_utils.cluster_command_call(
self,
cmd,
threads=self.cfg['max_threads'],
ram=2,
cfg=self.cfg,
err_log=self.output()['err_log'].path)
else:
pipeline_utils.command_call(cmd,
err_log=self.output()['err_log'].path)
def run(self):
cmd = [
'gatk4', '--java-options',
'"-Djava.io.tmpdir=%s"' % self.cfg['tmp_dir'], 'Mutect2', '-R',
self.cfg['fasta_file'], '--native-pair-hmm-threads',
self.cfg['max_threads'], '-I',
self.input()['preprocess']['bam'].path, '-O',
self.output()['mutect2_normal'].path
]
if self.cfg['cluster_exec']:
pipeline_utils.cluster_command_call(
self,
cmd,
threads=self.cfg['max_threads'],
ram=16,
cfg=self.cfg,
err_log=self.output()['err_log'].path)
else:
pipeline_utils.command_call(cmd,
err_log=self.output()['err_log'].path)
def run(self):
file_base = self.output()['filter_mutect2'].path.split('.vcf.gz')[0]
filtermutect = file_base + '.FilterMutectCalls.vcf'
raw_snps = file_base + '.snps.raw.vcf'
raw_indels = file_base + '.indels.raw.vcf'
filtered_snps = file_base + '.snps.filtered.vcf'
filtered_indels = file_base + '.indels.filtered.vcf'
# use default FilterMutectCalls first
cmd = [
'gatk4', '--java-options',
'"-Djava.io.tmpdir=%s"' % self.cfg['tmp_dir'], 'FilterMutectCalls',
'-R', self.cfg['fasta_file'], '-V',
self.input()['mutect2']['mutect2'].path, '-O', filtermutect
]
if self.cfg['cluster_exec']:
pipeline_utils.cluster_command_call(
self,
cmd,
threads=self.cfg['max_threads'],
ram=16,
cfg=self.cfg,
err_log=self.output()['err_log'].path)
else:
pipeline_utils.command_call(cmd,
err_log=self.output()['err_log'].path)
# select snps
cmd = [
'gatk4', '--java-options',
'"-Djava.io.tmpdir=%s"' % self.cfg['tmp_dir'], 'SelectVariants',
'-R', self.cfg['fasta_file'], '-V', filtermutect,
'--select-type-to-include', 'SNP', '-O', raw_snps
]
if self.cfg['cluster_exec']:
pipeline_utils.cluster_command_call(
self,
cmd,
threads=self.cfg['max_threads'],
ram=16,
cfg=self.cfg)
else:
pipeline_utils.command_call(cmd)
# select indels
cmd = [
'gatk4', '--java-options',
'"-Djava.io.tmpdir=%s"' % self.cfg['tmp_dir'], 'SelectVariants',
'-R', self.cfg['fasta_file'], '-V', filtermutect,
'--select-type-to-include', 'INDEL', '-O', raw_indels
]
if self.cfg['cluster_exec']:
pipeline_utils.cluster_command_call(
self,
cmd,
threads=self.cfg['max_threads'],
ram=16,
cfg=self.cfg)
else:
pipeline_utils.command_call(cmd)
# filter snps
cmd = [
'gatk4', '--java-options',
'"-Djava.io.tmpdir=%s"' % self.cfg['tmp_dir'], 'VariantFiltration',
'-R', self.cfg['fasta_file'], '-V', raw_snps, '-O', filtered_snps
]
for filter_name, filter_expression in [
('"FS60"', '"FS > 60.0"'), ('"MQ40"', '"MQ < 40.0"'),
('"MQRS-12.5"', '"MQRankSum < -12.5"'),
('"RPRS-8"', '"ReadPosRankSum < -8.0"'), ('"SOR3"', '"SOR > 3.0"'),
('"FS60"', '"FS > 60.0"')
]: # , ('"Q30"', '"QUAL < 30.0"')
cmd += [
'--filter-name', filter_name, '--filter-expression',
filter_expression
]
if not 'N' in self.cfg['cases'][self.case]:
cmd += [
'--filter-name', '"TLOD10"', '--filter-expression',
'"TLOD < 10.0"', '--filter-name', '"QD5"',
'--filter-expression', '"QD < 5.0"'
]
def run(self):
file_base = self.output()['filter_germline'].path.split('.vcf.gz')[0]
raw_snps = file_base + '.snps.raw.vcf'
raw_indels = file_base + '.indels.raw.vcf'
filtered_snps = file_base + '.snps.filtered.vcf'
filtered_indels = file_base + '.indels.filtered.vcf'
# select snps
cmd = [
'gatk4', '--java-options',
'"-Djava.io.tmpdir=%s"' % self.cfg['tmp_dir'], 'SelectVariants',
'-R', self.cfg['fasta_file'], '-V',
self.input()['haplotype_caller']['haplotype_caller'].path,
'--select-type-to-include', 'SNP', '-O', raw_snps
]
if self.cfg['cluster_exec']:
pipeline_utils.cluster_command_call(
self,
cmd,
threads=self.cfg['max_threads'],
ram=16,
cfg=self.cfg)
else:
pipeline_utils.command_call(cmd)
# select indels
cmd = [
'gatk4', '--java-options',
'"-Djava.io.tmpdir=%s"' % self.cfg['tmp_dir'], 'SelectVariants',
'-R', self.cfg['fasta_file'], '-V',
self.input()['haplotype_caller']['haplotype_caller'].path,
'--select-type-to-include', 'INDEL', '-O', raw_indels
]
if self.cfg['cluster_exec']:
pipeline_utils.cluster_command_call(
self,
cmd,
threads=self.cfg['max_threads'],
ram=16,
cfg=self.cfg)
else:
pipeline_utils.command_call(cmd)
# filter snps
cmd = [
'gatk4', '--java-options',
'"-Djava.io.tmpdir=%s"' % self.cfg['tmp_dir'], 'VariantFiltration',
'-R', self.cfg['fasta_file'], '-V', raw_snps, '-O', filtered_snps
]
for filter_name, filter_expression in [
('"filter_1"', '"QD < 2.0"'), ('"filter_2"', '"FS > 60.0"'),
('"filter_3"', '"MQ < 40.0"'),
('"filter_4"', '"MQRankSum < -12.5"'),
('"filter_5"', '"ReadPosRankSum < -8.0"')
]:
cmd += [
'--filter-name', filter_name, '--filter-expression',
filter_expression
]
if self.cfg['cluster_exec']:
pipeline_utils.cluster_command_call(
self,
cmd,
threads=self.cfg['max_threads'],
ram=16,
cfg=self.cfg)
else:
pipeline_utils.command_call(cmd)
# filter indels
cmd = [
'gatk4', '--java-options',
'"-Djava.io.tmpdir=%s"' % self.cfg['tmp_dir'], 'VariantFiltration',
'-R', self.cfg['fasta_file'], '-V', raw_indels, '-O',
filtered_indels
]
for filter_name, filter_expression in [('"filter_1"', '"QD < 2.0"'),
('"filter_2"', '"FS > 200.0"'),
('"filter_3"',
'"ReadPosRankSum < -20.0"')]:
cmd += [
'--filter-name', filter_name, '--filter-expression',
filter_expression
]
if self.cfg['cluster_exec']:
pipeline_utils.cluster_command_call(
self,
cmd,
threads=self.cfg['max_threads'],
ram=16,
cfg=self.cfg)
else:
pipeline_utils.command_call(cmd)
# combine snps and indels
cmd = [
'java',
'-Djava.io.tmpdir=%s' % self.cfg['tmp_dir'], '-jar', '$PICARD',
'MergeVcfs',
'I=%s' % filtered_snps,
'I=%s' % filtered_indels,
'O=%s' % self.output()['filter_germline'].path
]
if self.cfg['cluster_exec']:
pipeline_utils.cluster_command_call(
self,
cmd,
threads=self.cfg['max_threads'],
ram=16,
cfg=self.cfg,
err_log=self.output()['err_log'].path)
else:
pipeline_utils.command_call(cmd,
err_log=self.output()['err_log'].path)
for file in [raw_snps, raw_indels, filtered_snps, filtered_indels]:
try:
os.remove(file)
os.remove(file + '.idx')
except:
pass
def run(self): cmd = ['java', '-Djava.io.tmpdir=%s' % self.cfg['tmp_dir'], '-jar', '$PICARD', 'MarkDuplicates', 'I=%s' % self.input()['merge_bams']['merge_bams'].path, 'O=%s' % self.output()['mark_duplicates']['bam'].path, 'M=%s' % self.output()['mark_duplicates']['metrics'].path, 'TAGGING_POLICY=All'] if self.cfg['cluster_exec']: pipeline_utils.cluster_command_call(self, cmd, threads=self.cfg['max_threads'], ram=5, cfg=self.cfg, err_log=self.output()['err_log'].path) else: pipeline_utils.command_call(cmd, err_log=self.output()['err_log'].path)
def run(self): cmd = ['samtools', 'index', self.input()['mark_duplicates']['mark_duplicates']['bam'].path] if self.cfg['cluster_exec']: pipeline_utils.cluster_command_call(self, cmd, threads=1, ram=5, cfg=self.cfg, err_log=self.output()['err_log'].path) else: pipeline_utils.command_call(cmd, err_log=self.output()['err_log'].path)
def run(self): cmd = ['java', '-Djava.io.tmpdir=%s' % self.cfg['tmp_dir'], '-jar', '$GATK3', '-T', 'PrintReads', '-I', self.input()['mark_duplicates']['mark_duplicates']['bam'].path, '-R', self.cfg['fasta_file'], '-BQSR', self.input()['base_recalibrator']['base_recalibrator'].path, '-o', self.output()['apply_bqsr'].path] # self.input()['indel_realigner']['indel_realigner'][self.case][self.sample].path if self.cfg['cluster_exec']: pipeline_utils.cluster_command_call(self, cmd, threads=1, ram=5, cfg=self.cfg, err_log=self.output()['err_log'].path) else: pipeline_utils.command_call(cmd, err_log=self.output()['err_log'].path)
class vcf2maf_tumor(luigi.Task):
priority = 85
# resources = {'threads': 1}
cfg = luigi.DictParameter()
case = luigi.Parameter()
@property # This is necessary to assign a dynamic value to the 'threads' resource within a task
def resources(self):
return {'threads': self.cfg['max_threads']}
def requires(self):
return {'fpfilter': fpfilter(case=self.case, cfg=self.cfg)}
def output(self):
outputs = {
'vcf2maf':
luigi.LocalTarget(
os.path.join(self.cfg['output_dir'], self.case, 'variants',
'%s.maf' % self.case)),
'err_log':
luigi.LocalTarget(
os.path.join(self.cfg['output_dir'], self.case, 'log',
'%s_vcf2maf_err.txt' % self.case))
}
for task in outputs:
if isinstance(outputs[task], luigi.LocalTarget):
pipeline_utils.confirm_path(outputs[task].path)
return outputs
def run(self):
# if self.input()['filter_mutect2']['filter_mutect2'].path.endswith('.gz'):
# input_vcf = self.input()['filter_mutect2']['filter_mutect2'].path.split('.gz')[0]
# with gzip.open(self.input()['filter_mutect2']['filter_mutect2'].path, 'rb') as vcf_in, open(input_vcf, 'wb') as vcf_out:
# shutil.copyfileobj(vcf_in, vcf_out)
# else:
# input_vcf = self.input()['filter_mutect2']['filter_mutect2'].path.endswith('.gz')
cmd = [
'perl', '/root/pipeline/code/source/vcf2maf/vcf2maf.pl',
'--ref-fasta', self.cfg['fasta_file'], '--vep-forks',
self.cfg['max_threads'], '--input-vcf',
self.input()['fpfilter']['fpfilter'].path, '--output-maf',
self.output()['vcf2maf'].path, '--tumor-id',
'%s_T' % self.case
]
if 'N' in self.cfg['cases'][self.case]:
cmd += ['--normal-id', '%s_N' % self.case]
if self.cfg['cluster_exec']:
pipeline_utils.cluster_command_call(
self,
cmd,
threads=self.cfg['max_threads'],
ram=16,
cfg=self.cfg,
err_log=self.output()['err_log'].path)
else:
pipeline_utils.command_call(cmd,
err_log=self.output()['err_log'].path)
try:
os.remove(
'%s.vep.vcf' %
self.input()['fpfilter']['fpfilter'].path.split('.vcf')[0])
except:
pass
def run(self): cmd = ['bwa', 'index', '-a', 'bwtsw', self.cfg['fasta_file']] if self.cfg['cluster_exec']: pipeline_utils.cluster_command_call(self, cmd, threads=1, ram=5, cfg=self.cfg) #, err_log=self.output()['err_log'].path) else: pipeline_utils.command_call(cmd) #, err_log=self.output()['err_log'].path)
self.cfg['library_bed'], '--af-of-alleles-not-in-resource',
'0.0000025', '-I',
self.input()['T']['preprocess']['bam'].path, '-O',
self.output()['mutect2'].path
]
if 'N' in self.cfg['cases'][self.case]:
cmd += [
'-I',
self.input()['N']['preprocess']['bam'].path, '-normal',
'%s_N' % self.case
]
if self.cfg['cluster_exec']:
pipeline_utils.cluster_command_call(
self,
cmd,
threads=self.cfg['max_threads'],
ram=16,
cfg=self.cfg,
err_log=self.output()['err_log'].path)
else:
pipeline_utils.command_call(cmd,
err_log=self.output()['err_log'].path)
class filter_mutect2(luigi.Task):
priority = 86
resources = {'threads': 1}
cfg = luigi.DictParameter()
case = luigi.Parameter()
return requirements
def output(self):
# return {'realigner_target': luigi.LocalTarget(os.path.join(self.cfg['output_dir'], self.case, 'preprocess', '%s_%s_realigner_targets.intervals' % (self.case, self.sample))), 'err_log': luigi.LocalTarget(os.path.join(self.cfg['output_dir'], self.case, 'log', '%s_%s_realigner_target_err.txt' % (self.case, self.sample)))}
return {'realigner_target': luigi.LocalTarget(os.path.join(self.cfg['output_dir'], 'all_samples', 'preprocess', 'all_samples_realigner_targets.intervals')), 'file_map': luigi.LocalTarget(os.path.join(self.cfg['output_dir'], 'all_samples', 'preprocess', 'all_samples_realigner.map')), 'err_log': luigi.LocalTarget(os.path.join(self.cfg['output_dir'], 'all_samples', 'log', 'realigner_target_err.txt'))}
def run(self):
cmd = ['java', '-Djava.io.tmpdir=%s' % self.cfg['tmp_dir'], '-jar', '$GATK3', '-T', 'RealignerTargetCreator', '-R', self.cfg['fasta_file'], '--known', self.cfg['germline_indels'], '-nt', self.cfg['global_max_threads'], '-o', self.output()['realigner_target'].path]
file_map = []
for case in self.input():
for sample in self.input()[case]:
filename = self.input()[case][sample]['mark_duplicates']['mark_duplicates']['bam'].path
realigned_filename = filename.split('marked_duplicates.bam')[0] + 'realigned.bam'
file_map.append('%s\t%s' % (os.path.basename(filename), realigned_filename))
cmd += ['-I', filename]
if self.cfg['cluster_exec']:
pipeline_utils.cluster_command_call(self, cmd, threads=self.cfg['global_max_threads'], ram=48, cfg=self.cfg, err_log=self.output()['err_log'].path)
else:
pipeline_utils.command_call(cmd, err_log=self.output()['err_log'].path)
pipeline_utils.confirm_path(self.output()['file_map'].path)
with open(self.output()['file_map'].path, 'w') as f:
f.write('\n'.join(file_map))
class indel_realigner(luigi.Task):
priority = 94
resources = {'threads': 1}
cfg = luigi.DictParameter()
# @property # This is necessary to assign a dynamic value to the 'threads' resource within a task
# def resources(self):
# return {'threads': self.cfg['global_max_threads']}