def run_phlat(job, fastqs, sample_type, univ_options, phlat_options):
"""
Run PHLAT on a pair of input fastqs of type `sample_type`.
:param list fastqs: List of input fastq files
:param str sample_type: Description of the sample type to inject into the file name.
:param dict univ_options: Dict of universal options used by almost all tools
:param dict phlat_options: Options specific to PHLAT
:return: fsID for the HLA haplotype called from teh input fastqs
:rtype: toil.fileStore.FileID
"""
job.fileStore.logToMaster('Running phlat on %s:%s' %
(univ_options['patient'], sample_type))
work_dir = os.getcwd()
input_files = {
'input_1.fastq': fastqs[0],
'input_2.fastq': fastqs[1],
'phlat_index.tar.gz': phlat_options['index']
}
input_files = get_files_from_filestore(job,
input_files,
work_dir,
docker=False)
# Handle gzipped files
gz = '.gz' if is_gzipfile(input_files['input_1.fastq']) else ''
if gz:
for read_file in 'input_1.fastq', 'input_2.fastq':
os.symlink(read_file, read_file + gz)
input_files[read_file + gz] = input_files[read_file] + gz
# Untar the index
input_files['phlat_index'] = untargz(input_files['phlat_index.tar.gz'],
work_dir)
input_files = {key: docker_path(path) for key, path in input_files.items()}
parameters = [
'-1',
input_files['input_1.fastq' + gz],
'-2',
input_files['input_2.fastq' + gz],
'-index',
input_files['phlat_index'],
'-b2url',
'/usr/local/bin/bowtie2',
'-tag',
sample_type,
'-e',
'/home/phlat-1.0', # Phlat directory home
'-o',
'/data', # Output directory
'-p',
str(phlat_options['n'])
] # Number of threads
docker_call(tool='phlat',
tool_parameters=parameters,
work_dir=work_dir,
dockerhub=univ_options['dockerhub'],
tool_version=phlat_options['version'])
output_file = job.fileStore.writeGlobalFile(''.join(
[work_dir, '/', sample_type, '_HLA.sum']))
return output_file
def run_bwa(job, fastqs, sample_type, univ_options, bwa_options):
"""
Align a pair of fastqs with bwa.
:param list fastqs: The input fastqs for alignment
:param str sample_type: Description of the sample to inject into the filename
:param dict univ_options: Dict of universal options used by almost all tools
:param dict bwa_options: Options specific to bwa
:return: fsID for the generated sam
:rtype: toil.fileStore.FileID
"""
work_dir = os.getcwd()
input_files = {
'dna_1.fastq': fastqs[0],
'dna_2.fastq': fastqs[1],
'bwa_index.tar.gz': bwa_options['index']
}
input_files = get_files_from_filestore(job,
input_files,
work_dir,
docker=False)
# Handle gzipped file
gz = '.gz' if is_gzipfile(input_files['dna_1.fastq']) else ''
if gz:
for read_file in 'dna_1.fastq', 'dna_2.fastq':
os.symlink(read_file, read_file + gz)
input_files[read_file + gz] = input_files[read_file] + gz
# Untar the index
input_files['bwa_index'] = untargz(input_files['bwa_index.tar.gz'],
work_dir)
input_files = {key: docker_path(path) for key, path in input_files.items()}
parameters = [
'mem',
'-t',
str(bwa_options['n']),
'-v',
'1', # Don't print INFO messages to the stderr
'/'.join([input_files['bwa_index'], univ_options['ref']]),
input_files['dna_1.fastq' + gz],
input_files['dna_2.fastq' + gz]
]
with open(''.join([work_dir, '/', sample_type, '.sam']), 'w') as samfile:
docker_call(tool='bwa',
tool_parameters=parameters,
work_dir=work_dir,
dockerhub=univ_options['dockerhub'],
outfile=samfile,
tool_version=bwa_options['version'])
# samfile.name retains the path info
output_file = job.fileStore.writeGlobalFile(samfile.name)
job.fileStore.logToMaster('Ran bwa on %s:%s successfully' %
(univ_options['patient'], sample_type))
return output_file
def run_bwa(job, fastqs, sample_type, univ_options, bwa_options):
"""
This module aligns the SAMPLE_TYPE dna fastqs to the reference
ARGUMENTS -- <ST> depicts the sample type. Substitute with 'tumor'/'normal'
1. fastqs: Dict of list of input WGS/WXS fastqs
fastqs
+- '<ST>_dna': [<JSid for 1.fastq> , <JSid for 2.fastq>]
2. sample_type: string of 'tumor_dna' or 'normal_dna'
3. univ_options: Dict of universal arguments used by almost all tools
univ_options
+- 'dockerhub': <dockerhub to use>
4. bwa_options: Dict of parameters specific to bwa
bwa_options
|- 'tool_index': <JSid for the bwa index tarball>
+- 'n': <number of threads to allocate>
RETURN VALUES
1. output_files: Dict of aligned bam + reference (nested return)
output_files
|- '<ST>_fix_pg_sorted.bam': <JSid>
+- '<ST>_fix_pg_sorted.bam.bai': <JSid>
This module corresponds to nodes 3 and 4 on the tree
"""
job.fileStore.logToMaster('Running bwa on %s:%s' % (univ_options['patient'], sample_type))
work_dir = os.getcwd()
input_files = {
'dna_1.fastq': fastqs[0],
'dna_2.fastq': fastqs[1],
'bwa_index.tar.gz': bwa_options['tool_index']}
input_files = get_files_from_filestore(job, input_files, work_dir, docker=False)
# Handle gzipped file
gz = '.gz' if is_gzipfile(input_files['dna_1.fastq']) else ''
if gz:
for read_file in 'dna_1.fastq', 'dna_2.fastq':
os.symlink(read_file, read_file + gz)
input_files[read_file + gz] = input_files[read_file] + gz
# Untar the index
input_files['bwa_index'] = untargz(input_files['bwa_index.tar.gz'], work_dir)
input_files = {key: docker_path(path) for key, path in input_files.items()}
parameters = ['mem',
'-t', str(bwa_options['n']),
'-v', '1', # Don't print INFO messages to the stderr
'/'.join([input_files['bwa_index'], 'hg19']),
input_files['dna_1.fastq' + gz],
input_files['dna_2.fastq' + gz]]
with open(''.join([work_dir, '/', sample_type, '_aligned.sam']), 'w') as samfile:
docker_call(tool='bwa', tool_parameters=parameters, work_dir=work_dir,
dockerhub=univ_options['dockerhub'], outfile=samfile)
# samfile.name retains the path info
output_file = job.fileStore.writeGlobalFile(samfile.name)
return output_file
def run_phlat(job, fastqs, sample_type, univ_options, phlat_options):
"""
This module will run PHLAT on SAMPLE_TYPE fastqs.
ARGUMENTS -- <ST> depicts the sample type. Substitute with 'tumor_dna',
'normal_dna', or 'tumor_rna'
1. fastqs: Dict of list of input WGS/WXS fastqs
fastqs
+- '<ST>': [<JSid for 1.fastq> , <JSid for 2.fastq>]
2. sample_type: string of 'tumor' or 'normal'
3. univ_options: Dict of universal arguments used by almost all tools
univ_options
+- 'dockerhub': <dockerhub to use>
4. phlat_options: Dict of parameters specific to phlat
phlat_options
|- 'tool_index': <JSid for the PHLAT index tarball>
+- 'n': <number of threads to allocate>
RETURN VALUES
1. output_file: <JSid for the allele predictions for ST>
This module corresponds to nodes 5, 6 and 7 on the tree
"""
job.fileStore.logToMaster('Running phlat on %s:%s' % (univ_options['patient'], sample_type))
print(phlat_options, file=sys.stderr)
work_dir = os.getcwd()
input_files = {
'input_1.fastq': fastqs[0],
'input_2.fastq': fastqs[1],
'phlat_index.tar.gz': phlat_options['tool_index']}
input_files = get_files_from_filestore(job, input_files, work_dir, docker=False)
# Handle gzipped files
gz = '.gz' if is_gzipfile(input_files['input_1.fastq']) else ''
if gz:
for read_file in 'input_1.fastq', 'input_2.fastq':
os.symlink(read_file, read_file + gz)
input_files[read_file + gz] = input_files[read_file] + gz
# Untar the index
input_files['phlat_index'] = untargz(input_files['phlat_index.tar.gz'], work_dir)
input_files = {key: docker_path(path) for key, path in input_files.items()}
parameters = ['-1', input_files['input_1.fastq' + gz],
'-2', input_files['input_2.fastq' + gz],
'-index', input_files['phlat_index'],
'-b2url', '/usr/local/bin/bowtie2',
'-tag', sample_type,
'-e', '/home/phlat-1.0', # Phlat directory home
'-o', '/data', # Output directory
'-p', str(phlat_options['n'])] # Number of threads
docker_call(tool='phlat', tool_parameters=parameters, work_dir=work_dir,
dockerhub=univ_options['dockerhub'])
output_file = job.fileStore.writeGlobalFile(''.join([work_dir, '/', sample_type, '_HLA.sum']))
return output_file
def run_phlat(job, fastqs, sample_type, univ_options, phlat_options):
"""
This module will run PHLAT on SAMPLE_TYPE fastqs.
ARGUMENTS -- <ST> depicts the sample type. Substitute with 'tumor_dna',
'normal_dna', or 'tumor_rna'
1. fastqs: Dict of list of input WGS/WXS fastqs
fastqs
+- '<ST>': [<JSid for 1.fastq> , <JSid for 2.fastq>]
2. sample_type: string of 'tumor' or 'normal'
3. univ_options: Dict of universal arguments used by almost all tools
univ_options
+- 'dockerhub': <dockerhub to use>
4. phlat_options: Dict of parameters specific to phlat
phlat_options
|- 'tool_index': <JSid for the PHLAT index tarball>
+- 'n': <number of threads to allocate>
RETURN VALUES
1. output_file: <JSid for the allele predictions for ST>
This module corresponds to nodes 5, 6 and 7 on the tree
"""
job.fileStore.logToMaster('Running phlat on %s:%s' % (univ_options['patient'], sample_type))
work_dir = os.getcwd()
input_files = {
'input_1.fastq': fastqs[0],
'input_2.fastq': fastqs[1],
'phlat_index.tar.gz': phlat_options['tool_index']}
input_files = get_files_from_filestore(job, input_files, work_dir, docker=False)
# Handle gzipped files
gz = '.gz' if is_gzipfile(input_files['input_1.fastq']) else ''
if gz:
for read_file in 'input_1.fastq', 'input_2.fastq':
os.symlink(read_file, read_file + gz)
input_files[read_file + gz] = input_files[read_file] + gz
# Untar the index
input_files['phlat_index'] = untargz(input_files['phlat_index.tar.gz'], work_dir)
input_files = {key: docker_path(path) for key, path in input_files.items()}
parameters = ['-1', input_files['input_1.fastq' + gz],
'-2', input_files['input_2.fastq' + gz],
'-index', input_files['phlat_index'],
'-b2url', '/usr/local/bin/bowtie2',
'-tag', sample_type,
'-e', '/home/phlat-1.0', # Phlat directory home
'-o', '/data', # Output directory
'-p', str(phlat_options['n'])] # Number of threads
docker_call(tool='phlat', tool_parameters=parameters, work_dir=work_dir,
dockerhub=univ_options['dockerhub'])
output_file = job.fileStore.writeGlobalFile(''.join([work_dir, '/', sample_type, '_HLA.sum']))
return output_file
def run_cutadapt(job, fastqs, univ_options, cutadapt_options):
"""
Runs cutadapt on the input RNA fastq files.
:param list fastqs: List of fsIDs for input an RNA-Seq fastq pair
:param dict univ_options: Dict of universal options used by almost all tools
:param dict cutadapt_options: Options specific to cutadapt
:return: List of fsIDs of cutadapted fastqs
:rtype: list[toil.fileStore.FileID]
"""
work_dir = os.getcwd()
input_files = {'rna_1.fastq': fastqs[0], 'rna_2.fastq': fastqs[1]}
input_files = get_files_from_filestore(job,
input_files,
work_dir,
docker=False)
# Handle gzipped file
gz = '.gz' if is_gzipfile(input_files['rna_1.fastq']) else ''
if gz:
for read_file in 'rna_1.fastq', 'rna_2.fastq':
os.symlink(read_file, read_file + gz)
input_files[read_file + gz] = input_files[read_file] + gz
input_files = {
key: docker_path(path)
for key, path in list(input_files.items())
}
parameters = [
'-a',
cutadapt_options['a'], # Fwd read 3' adapter
'-A',
cutadapt_options['A'], # Rev read 3' adapter
'-m',
'35', # Minimum size of read
'-o',
docker_path('rna_cutadapt_1.fastq.gz'), # Output for R1
'-p',
docker_path('rna_cutadapt_2.fastq.gz'), # Output for R2
input_files['rna_1.fastq' + gz],
input_files['rna_2.fastq' + gz]
]
docker_call(tool='cutadapt',
tool_parameters=parameters,
work_dir=work_dir,
dockerhub=univ_options['dockerhub'],
tool_version=cutadapt_options['version'])
output_files = []
for fastq_file in ['rna_cutadapt_1.fastq.gz', 'rna_cutadapt_2.fastq.gz']:
output_files.append(
job.fileStore.writeGlobalFile('/'.join([work_dir, fastq_file])))
job.fileStore.logToMaster('Ran cutadapt on %s successfully' %
univ_options['patient'])
return output_files
def get_patient_vcf(job, patient_dict):
"""
Convenience function to get the vcf from the patient dict
:param dict patient_dict: dict of patient info
:return: The vcf
:rtype: toil.fileStore.FileID
"""
temp = job.fileStore.readGlobalFile(patient_dict['mutation_vcf'],
os.path.join(os.getcwd(), 'temp.gz'))
if is_gzipfile(temp):
outfile = gunzip(temp)
job.fileStore.deleteGlobalFile(patient_dict['mutation_vcf'])
else:
outfile = patient_dict['mutation_vcf']
return outfile
def run_cutadapt(job, fastqs, univ_options, cutadapt_options):
"""
This module runs cutadapt on the input RNA fastq files and then calls the RNA aligners.
ARGUMENTS
1. fastqs: List of input RNA-Seq fastqs [<JSid for 1.fastq> , <JSid for 2.fastq>]
2. univ_options: Dict of universal arguments used by almost all tools
univ_options
+- 'dockerhub': <dockerhub to use>
3. cutadapt_options: Dict of parameters specific to cutadapt
cutadapt_options
|- 'a': <sequence of 3' adapter to trim from fwd read>
+- 'A': <sequence of 3' adapter to trim from rev read>
RETURN VALUES
1. output_files: Dict of cutadapted fastqs
output_files
|- 'rna_cutadapt_1.fastq': <JSid>
+- 'rna_cutadapt_2.fastq': <JSid>
This module corresponds to node 2 on the tree
"""
job.fileStore.logToMaster('Running cutadapt on %s' % univ_options['patient'])
work_dir = os.getcwd()
input_files = {
'rna_1.fastq': fastqs[0],
'rna_2.fastq': fastqs[1]}
input_files = get_files_from_filestore(job, input_files, work_dir, docker=False)
# Handle gzipped file
gz = '.gz' if is_gzipfile(input_files['rna_1.fastq']) else ''
if gz:
for read_file in 'rna_1.fastq', 'rna_2.fastq':
os.symlink(read_file, read_file + gz)
input_files[read_file + gz] = input_files[read_file] + gz
input_files = {key: docker_path(path) for key, path in input_files.items()}
parameters = ['-a', cutadapt_options['a'], # Fwd read 3' adapter
'-A', cutadapt_options['A'], # Rev read 3' adapter
'-m', '35', # Minimum size of read
'-o', docker_path('rna_cutadapt_1.fastq.gz'), # Output for R1
'-p', docker_path('rna_cutadapt_2.fastq.gz'), # Output for R2
input_files['rna_1.fastq' + gz],
input_files['rna_2.fastq' + gz]]
docker_call(tool='cutadapt', tool_parameters=parameters, work_dir=work_dir,
dockerhub=univ_options['dockerhub'])
output_files = []
for fastq_file in ['rna_cutadapt_1.fastq.gz', 'rna_cutadapt_2.fastq.gz']:
output_files.append(job.fileStore.writeGlobalFile('/'.join([work_dir, fastq_file])))
return output_files
def run_cutadapt(job, fastqs, univ_options, cutadapt_options):
"""
This module runs cutadapt on the input RNA fastq files and then calls the RNA aligners.
ARGUMENTS
1. fastqs: List of input RNA-Seq fastqs [<JSid for 1.fastq> , <JSid for 2.fastq>]
2. univ_options: Dict of universal arguments used by almost all tools
univ_options
+- 'dockerhub': <dockerhub to use>
3. cutadapt_options: Dict of parameters specific to cutadapt
cutadapt_options
|- 'a': <sequence of 3' adapter to trim from fwd read>
+- 'A': <sequence of 3' adapter to trim from rev read>
RETURN VALUES
1. output_files: Dict of cutadapted fastqs
output_files
|- 'rna_cutadapt_1.fastq': <JSid>
+- 'rna_cutadapt_2.fastq': <JSid>
This module corresponds to node 2 on the tree
"""
job.fileStore.logToMaster('Running cutadapt on %s' % univ_options['patient'])
work_dir = os.getcwd()
input_files = {
'rna_1.fastq': fastqs[0],
'rna_2.fastq': fastqs[1]}
input_files = get_files_from_filestore(job, input_files, work_dir, docker=False)
# Handle gzipped file
gz = '.gz' if is_gzipfile(input_files['rna_1.fastq']) else ''
if gz:
for read_file in 'rna_1.fastq', 'rna_2.fastq':
os.symlink(read_file, read_file + gz)
input_files[read_file + gz] = input_files[read_file] + gz
input_files = {key: docker_path(path) for key, path in input_files.items()}
parameters = ['-a', cutadapt_options['a'], # Fwd read 3' adapter
'-A', cutadapt_options['A'], # Rev read 3' adapter
'-m', '35', # Minimum size of read
'-o', docker_path('rna_cutadapt_1.fastq.gz'), # Output for R1
'-p', docker_path('rna_cutadapt_2.fastq.gz'), # Output for R2
input_files['rna_1.fastq' + gz],
input_files['rna_2.fastq' + gz]]
docker_call(tool='cutadapt', tool_parameters=parameters, work_dir=work_dir,
dockerhub=univ_options['dockerhub'])
output_files = []
for fastq_file in ['rna_cutadapt_1.fastq.gz', 'rna_cutadapt_2.fastq.gz']:
output_files.append(job.fileStore.writeGlobalFile('/'.join([work_dir, fastq_file])))
return output_files
def run_bwa(job, fastqs, sample_type, univ_options, bwa_options):
"""
This module aligns the SAMPLE_TYPE dna fastqs to the reference
ARGUMENTS -- <ST> depicts the sample type. Substitute with 'tumor'/'normal'
1. fastqs: Dict of list of input WGS/WXS fastqs
fastqs
+- '<ST>_dna': [<JSid for 1.fastq> , <JSid for 2.fastq>]
2. sample_type: string of 'tumor_dna' or 'normal_dna'
3. univ_options: Dict of universal arguments used by almost all tools
univ_options
+- 'dockerhub': <dockerhub to use>
4. bwa_options: Dict of parameters specific to bwa
bwa_options
|- 'tool_index': <JSid for the bwa index tarball>
+- 'n': <number of threads to allocate>
RETURN VALUES
1. output_files: Dict of aligned bam + reference (nested return)
output_files
|- '<ST>_fix_pg_sorted.bam': <JSid>
+- '<ST>_fix_pg_sorted.bam.bai': <JSid>
This module corresponds to nodes 3 and 4 on the tree
"""
job.fileStore.logToMaster('Running bwa on %s:%s' %
(univ_options['patient'], sample_type))
work_dir = os.getcwd()
input_files = {
'dna_1.fastq': fastqs[0],
'dna_2.fastq': fastqs[1],
'bwa_index.tar.gz': bwa_options['tool_index']
}
input_files = get_files_from_filestore(job,
input_files,
work_dir,
docker=False)
# Handle gzipped file
gz = '.gz' if is_gzipfile(input_files['dna_1.fastq']) else ''
if gz:
for read_file in 'dna_1.fastq', 'dna_2.fastq':
os.symlink(read_file, read_file + gz)
input_files[read_file + gz] = input_files[read_file] + gz
# Untar the index
input_files['bwa_index'] = untargz(input_files['bwa_index.tar.gz'],
work_dir)
input_files = {key: docker_path(path) for key, path in input_files.items()}
parameters = [
'mem',
'-t',
str(bwa_options['n']),
'-v',
'1', # Don't print INFO messages to the stderr
'/'.join([input_files['bwa_index'], 'hg19']),
input_files['dna_1.fastq' + gz],
input_files['dna_2.fastq' + gz]
]
with open(''.join([work_dir, '/', sample_type, '_aligned.sam']),
'w') as samfile:
docker_call(tool='bwa',
tool_parameters=parameters,
work_dir=work_dir,
dockerhub=univ_options['dockerhub'],
outfile=samfile)
# samfile.name retains the path info
output_file = job.fileStore.writeGlobalFile(samfile.name)
return output_file
def run_star(job, fastqs, univ_options, star_options):
"""
Align a pair of fastqs with STAR.
:param list fastqs: The input fastqs for alignment
:param dict univ_options: Dict of universal options used by almost all tools
:param dict star_options: Options specific to star
:return: Dict containing output genome bam, genome bai, and transcriptome bam
output_files:
|- 'rnaAligned.toTranscriptome.out.bam': fsID
+- 'rnaAligned.out.bam': fsID
+- 'rnaChimeric.out.junction': fsID
:rtype: dict
"""
assert star_options['type'] in ('star', 'starlong')
work_dir = os.getcwd()
input_files = {
'rna_cutadapt_1.fastq': fastqs[0],
'rna_cutadapt_2.fastq': fastqs[1],
'star_index.tar.gz': star_options['index']}
input_files = get_files_from_filestore(job, input_files, work_dir,
docker=False)
# Handle gzipped file
gz = '.gz' if is_gzipfile(input_files['rna_cutadapt_1.fastq']) else ''
if gz:
for read_file in 'rna_cutadapt_1.fastq', 'rna_cutadapt_2.fastq':
os.symlink(read_file, read_file + gz)
input_files[read_file + gz] = input_files[read_file] + gz
# Untar the index
input_files['star_index'] = untargz(input_files['star_index.tar.gz'], work_dir)
# Check to see if user is using a STAR-Fusion index
star_fusion_idx = os.path.join(input_files['star_index'], 'ref_genome.fa.star.idx')
if os.path.exists(star_fusion_idx):
input_files['star_index'] = star_fusion_idx
input_files = {key: docker_path(path, work_dir=work_dir) for key, path in input_files.items()}
# Using recommended STAR-Fusion parameters:
# https://github.com/STAR-Fusion/STAR-Fusion/wiki
parameters = ['--runThreadN', str(star_options['n']),
'--genomeDir', input_files['star_index'],
'--twopassMode', 'Basic',
'--outReadsUnmapped', 'None',
'--chimSegmentMin', '12',
'--chimJunctionOverhangMin', '12',
'--alignSJDBoverhangMin', '10',
'--alignMatesGapMax', '200000',
'--alignIntronMax', '200000',
'--chimSegmentReadGapMax', 'parameter', '3',
'--alignSJstitchMismatchNmax', '5', '-1', '5', '5',
'--outFileNamePrefix', 'rna',
'--readFilesIn',
input_files['rna_cutadapt_1.fastq' + gz],
input_files['rna_cutadapt_2.fastq' + gz],
'--outSAMattributes', 'NH', 'HI', 'AS', 'NM', 'MD',
'--outSAMtype', 'BAM', 'Unsorted',
'--quantMode', 'TranscriptomeSAM']
if gz:
parameters.extend(['--readFilesCommand', 'zcat'])
if star_options['type'] == 'star':
docker_call(tool='star', tool_parameters=parameters, work_dir=work_dir,
dockerhub=univ_options['dockerhub'], tool_version=star_options['version'])
else:
docker_call(tool='starlong', tool_parameters=parameters, work_dir=work_dir,
dockerhub=univ_options['dockerhub'], tool_version=star_options['version'])
output_files = defaultdict()
for output_file in ['rnaAligned.toTranscriptome.out.bam',
'rnaAligned.out.bam',
'rnaChimeric.out.junction']:
output_files[output_file] = job.fileStore.writeGlobalFile('/'.join([work_dir, output_file]))
export_results(job, output_files['rnaAligned.toTranscriptome.out.bam'], 'rna_transcriptome.bam',
univ_options, subfolder='alignments')
export_results(job, output_files['rnaChimeric.out.junction'], 'rna_chimeric.junction',
univ_options, subfolder='mutations/fusions')
job.fileStore.logToMaster('Ran STAR on %s successfully' % univ_options['patient'])
return output_files
def run_star(job, fastqs, univ_options, star_options):
"""
This module uses STAR to align the RNA fastqs to the reference
ARGUMENTS
1. fastqs: REFER RETURN VALUE of run_cutadapt()
2. univ_options: Dict of universal arguments used by almost all tools
univ_options
+- 'dockerhub': <dockerhub to use>
3. star_options: Dict of parameters specific to STAR
star_options
|- 'tool_index': <JSid for the STAR index tarball>
+- 'n': <number of threads to allocate>
RETURN VALUES
1. output_files: Dict of aligned bams
output_files
|- 'rnaAligned.toTranscriptome.out.bam': <JSid>
+- 'rnaAligned.sortedByCoord.out.bam': Dict of genome bam + bai
|- 'rna_fix_pg_sorted.bam': <JSid>
+- 'rna_fix_pg_sorted.bam.bai': <JSid>
This module corresponds to node 9 on the tree
"""
assert star_options['type'] in ('star', 'starlong')
job.fileStore.logToMaster('Running STAR on %s' % univ_options['patient'])
work_dir = os.getcwd()
input_files = {
'rna_cutadapt_1.fastq': fastqs[0],
'rna_cutadapt_2.fastq': fastqs[1],
'star_index.tar.gz': star_options['tool_index']}
input_files = get_files_from_filestore(job, input_files, work_dir,
docker=False)
# Handle gzipped file
gz = '.gz' if is_gzipfile(input_files['rna_cutadapt_1.fastq']) else ''
if gz:
for read_file in 'rna_cutadapt_1.fastq', 'rna_cutadapt_2.fastq':
os.symlink(read_file, read_file + gz)
input_files[read_file + gz] = input_files[read_file] + gz
# Untar the index
input_files['star_index'] = untargz(input_files['star_index.tar.gz'], work_dir)
input_files = {key: docker_path(path) for key, path in input_files.items()}
parameters = ['--runThreadN', str(star_options['n']),
'--genomeDir', input_files['star_index'],
'--outFileNamePrefix', 'rna',
'--readFilesIn',
input_files['rna_cutadapt_1.fastq' + gz],
input_files['rna_cutadapt_2.fastq' + gz],
'--outSAMattributes', 'NH', 'HI', 'AS', 'NM', 'MD',
'--outSAMtype', 'BAM', 'SortedByCoordinate',
'--quantMode', 'TranscriptomeSAM']
if gz:
parameters.extend(['--readFilesCommand', 'zcat'])
if star_options['type'] == 'star':
docker_call(tool='star', tool_parameters=parameters, work_dir=work_dir,
dockerhub=univ_options['dockerhub'])
else:
docker_call(tool='starlong', tool_parameters=parameters, work_dir=work_dir,
dockerhub=univ_options['dockerhub'])
output_files = defaultdict()
for bam_file in ['rnaAligned.toTranscriptome.out.bam',
'rnaAligned.sortedByCoord.out.bam']:
output_files[bam_file] = job.fileStore.writeGlobalFile('/'.join([
work_dir, bam_file]))
return output_files
def run_star(job, fastqs, univ_options, star_options):
"""
This module uses STAR to align the RNA fastqs to the reference
ARGUMENTS
1. fastqs: REFER RETURN VALUE of run_cutadapt()
2. univ_options: Dict of universal arguments used by almost all tools
univ_options
+- 'dockerhub': <dockerhub to use>
3. star_options: Dict of parameters specific to STAR
star_options
|- 'tool_index': <JSid for the STAR index tarball>
+- 'n': <number of threads to allocate>
RETURN VALUES
1. output_files: Dict of aligned bams
output_files
|- 'rnaAligned.toTranscriptome.out.bam': <JSid>
+- 'rnaAligned.sortedByCoord.out.bam': Dict of genome bam + bai
|- 'rna_fix_pg_sorted.bam': <JSid>
+- 'rna_fix_pg_sorted.bam.bai': <JSid>
This module corresponds to node 9 on the tree
"""
assert star_options['type'] in ('star', 'starlong')
job.fileStore.logToMaster('Running STAR on %s' % univ_options['patient'])
work_dir = os.getcwd()
input_files = {
'rna_cutadapt_1.fastq': fastqs[0],
'rna_cutadapt_2.fastq': fastqs[1],
'star_index.tar.gz': star_options['tool_index']
}
input_files = get_files_from_filestore(job,
input_files,
work_dir,
docker=False)
# Handle gzipped file
gz = '.gz' if is_gzipfile(input_files['rna_cutadapt_1.fastq']) else ''
if gz:
for read_file in 'rna_cutadapt_1.fastq', 'rna_cutadapt_2.fastq':
os.symlink(read_file, read_file + gz)
input_files[read_file + gz] = input_files[read_file] + gz
# Untar the index
input_files['star_index'] = untargz(input_files['star_index.tar.gz'],
work_dir)
input_files = {key: docker_path(path) for key, path in input_files.items()}
parameters = [
'--runThreadN',
str(star_options['n']), '--genomeDir', input_files['star_index'],
'--outFileNamePrefix', 'rna', '--readFilesIn',
input_files['rna_cutadapt_1.fastq' + gz],
input_files['rna_cutadapt_2.fastq' + gz], '--outSAMattributes', 'NH',
'HI', 'AS', 'NM', 'MD', '--outSAMtype', 'BAM', 'SortedByCoordinate',
'--quantMode', 'TranscriptomeSAM'
]
if gz:
parameters.extend(['--readFilesCommand', 'zcat'])
if star_options['type'] == 'star':
docker_call(tool='star',
tool_parameters=parameters,
work_dir=work_dir,
dockerhub=univ_options['dockerhub'])
else:
docker_call(tool='starlong',
tool_parameters=parameters,
work_dir=work_dir,
dockerhub=univ_options['dockerhub'])
output_files = defaultdict()
for bam_file in [
'rnaAligned.toTranscriptome.out.bam',
'rnaAligned.sortedByCoord.out.bam'
]:
output_files[bam_file] = job.fileStore.writeGlobalFile('/'.join(
[work_dir, bam_file]))
return output_files
