def extract_bam_unmap(self):
jobs = []
for sample in self.samples:
sclip_directory = os.path.join("sclip", sample.name)
sclip_file_prefix = os.path.join("sclip", sample.name, sample.name + ".")
extract_directory = os.path.join("extract", sample.name)
extract_file_prefix = os.path.join("extract", sample.name, sample.name + ".")
jobMkdir = Job(command="if [ ! -d " + extract_directory + " ]; then mkdir -p " + extract_directory + "; fi")
## extract Orphan
job = concat_jobs([
jobMkdir,
concat_jobs([
samtools.view(sclip_file_prefix + "scOthers.bam", extract_file_prefix + "ORPHAN.bam", "-b -h -f 12 -F 256"),
samtools.sort(extract_file_prefix + "ORPHAN.bam", extract_file_prefix + "ORPHAN.sName", True)
])
], name="extract_bam_ORPHAN_" + sample.name)
jobs.append(job)
## extract OEA close to sclip
job = concat_jobs([
jobMkdir,
concat_jobs([
samtools.view(sclip_file_prefix + "sc.bam", extract_file_prefix + "OEAUNMAP.1.bam", "-b -h -f 68 -F 264"),
samtools.sort(extract_file_prefix + "OEAUNMAP.1.bam", extract_file_prefix + "OEAUNMAP.1.sName", True)
])
], name="extract_bam_OEAUNMAP1_" + sample.name)
jobs.append(job)
job = concat_jobs([
jobMkdir,
concat_jobs([
samtools.view(sclip_file_prefix + "sc.bam", extract_file_prefix + "OEAUNMAP.2.bam", "-b -h -f 132 -F 264"),
samtools.sort(extract_file_prefix + "OEAUNMAP.2.bam", extract_file_prefix + "OEAUNMAP.2.sName", True)
])
], name="extract_bam_OEAUNMAP2_" + sample.name)
jobs.append(job)
job = concat_jobs([
jobMkdir,
concat_jobs([
samtools.view(sclip_file_prefix + "sc.bam", extract_file_prefix + "OEAMAP.bam", "-b -h -f 8 -F 1284"),
samtools.sort(extract_file_prefix + "OEAMAP.bam", extract_file_prefix + "OEAMAP.sName", True)
])
], name="extract_bam_OEAMAP_" + sample.name)
jobs.append(job)
return jobs
def samtools_bam_sort(self):
"""
Sorts bam by readname prior to picard_sam_to_fastq step in order to minimize memory consumption.
If bam file is small and the memory requirements are reasonable, this step can be skipped.
"""
jobs = []
for readset in self.readsets:
# If readset FASTQ files are available, skip this step
if not readset.fastq1:
if readset.bam:
sortedBamPrefix = re.sub("\.bam$", ".sorted",
readset.bam.strip())
job = samtools.sort(readset.bam,
sortedBamPrefix,
sort_by_name=True)
job.name = "samtools_bam_sort." + readset.name
job.removable_files = [sortedBamPrefix + ".bam"]
job.samples = [readset.sample]
jobs.append(job)
else:
raise Exception(
"Error: BAM file not available for readset \"" +
readset.name + "\"!")
return jobs
def create_hic_file(self):
"""
A .hic file is created per sample in order to visualize in JuiceBox, WashU epigenome browser or as input for other tools.
For more detailed information about the JuiceBox visit: [JuiceBox] (http://www.aidenlab.org/software.html)
"""
jobs = []
for sample in self.samples:
sample_input = os.path.join(self.output_dirs['bams_output_directory'], sample.name, sample.name + ".merged.bam")
sortedBamPrefix = re.sub("\.merged.bam", ".merged.sorted", sample_input.strip())
sortedBam = sortedBamPrefix + ".bam"
hic_output = os.path.join(self.output_dirs['hicfiles_output_directory'], sample.name + ".hic")
job = concat_jobs([
Job(command="mkdir -p " + self.output_dirs['hicfiles_output_directory']),
samtools.sort(sample_input, sortedBamPrefix, sort_by_name=True),
hic.create_input(sortedBam, sample.name),
hic.create_hic(sample.name + ".juicebox.input.sorted", hic_output, self.genome)
])
job.name = "create_hic_file." + sample.name
job.samples = [sample]
jobs.append(job)
return jobs
def create_hic_file(self):
"""
A .hic file is created per sample in order to visualize in JuiceBox, WashU epigenome browser or as input for other tools.
For more detailed information about the JuiceBox visit: [JuiceBox] (http://www.aidenlab.org/software.html)
"""
jobs = []
for sample in self.samples:
sample_input = os.path.join(
self.output_dirs['bams_output_directory'], sample.name,
sample.name + ".merged.bam")
sortedBamPrefix = re.sub("\.merged.bam", ".merged.sorted",
sample_input.strip())
sortedBam = sortedBamPrefix + ".bam"
hic_output = os.path.join(
self.output_dirs['hicfiles_output_directory'],
sample.name + ".hic")
command_sort = samtools.sort(sample_input,
sortedBamPrefix,
sort_by_name=True)
command_input = Job(
input_files=[sortedBam],
output_files=[
sample.name + ".juicebox.input",
sample.name + ".juicebox.input.sorted"
],
module_entries=[["create_hic_file", "module_mugqic_tools"]],
name="create_hic_file." + sample.name,
command="bash {CreateHicFileInput} {sortedBam} {name} {tmpDir}"
.format(CreateHicFileInput='CreateHicFileInput.sh',
sortedBam=sortedBam,
name=sample.name,
tmpDir=os.path.expandvars("$(pwd)")),
removable_files=[
sample.name + ".juicebox.input",
sample.name + ".juicebox.input.sorted", sortedBam
])
command_juicebox = Job(
input_files=[sample.name + ".juicebox.input.sorted"],
output_files=[hic_output],
module_entries=[["create_hic_file", "module_java"]],
name="create_hic_file." + sample.name,
command=
"mkdir -p {hic_output} && java -jar {juicer} pre -q {q} {name} {output} {assembly}"
.format(
hic_output=self.output_dirs['hicfiles_output_directory'],
juicer=os.path.expandvars(
config.param('create_hic_file', 'JuicerPath')),
q=config.param('create_hic_file', 'q'),
name=sample.name + ".juicebox.input.sorted",
output=hic_output,
assembly=self.genome))
job = concat_jobs([command_sort, command_input, command_juicebox])
job.name = "create_hic_file." + sample.name
jobs.append(job)
return jobs