def bait_intersect(self):
"""
provided with a bed file, for example a bed of GWAS snps or features of interest, this method returns the lines in the bed file that intersect with the baits that have significant interactions.
Input bed must have 4 columns (<chr> <start> <end> <annotation>) and must be tab separated.
"""
jobs = []
chicago_output_dir = self.output_dirs['chicago_output_directory']
intersect_output_dir = self.output_dirs['intersect_ouput_directory']
other_options = config.param('bait_intersect', 'other_options', required = False)
features_file = config.param('bait_intersect', 'features_file', required = True)
sorted_features_file = os.path.splitext(features_file)[0] + ".sorted.bed"
output_dir = os.path.join(chicago_output_dir, intersect_output_dir)
if features_file != "None":
job_create_dir = Job(command = "mkdir -p {output_dir}".format(output_dir = output_dir))
job_sort_features_file = bedops.sort_bed(features_file, sorted_features_file)
job = concat_jobs([
job_create_dir,
job_sort_features_file
])
job.name = "bait_intersect.sort_feature." + os.path.basename(sorted_features_file)
job.removable_files = [sorted_features_file]
job.samples = self.samples
jobs.append(job)
for sample in self.samples:
sample_output_dir = os.path.join(chicago_output_dir, sample.name, "data")
ibed_file = os.path.join(sample_output_dir, sample.name + ".ibed")
sorted_ibed_file = re.sub("\.ibed$", ".bait.sorted.bed", ibed_file)
output_file_prefix = os.path.join(output_dir, os.path.splitext(os.path.basename(features_file))[0] + "_" + os.path.splitext(os.path.basename(ibed_file))[0])
job_extract_bait_bed = tools.sh_extract_bait_bed(ibed_file, sample.name)
job_sort_ibed = bedops.sort_bed(ibed_file + ".bait", sorted_ibed_file)
job_intersect = bedtools.intersect_beds(sorted_features_file, sorted_ibed_file, output_file_prefix + ".tmp", "-wa -u")
job_bedopsMap = bedops.bedmap_echoMapId(output_file_prefix + ".tmp", sorted_ibed_file, output_file_prefix + ".bait_intersect.bed")
job = concat_jobs([
job_extract_bait_bed,
job_sort_ibed,
job_intersect,
job_bedopsMap
])
job.name = "bait_intersect." + sample.name
job.removable_files = [sorted_ibed_file, sorted_features_file, output_file_prefix + ".tmp", ibed_file + ".bait"]
job.samples = [sample]
jobs.append(job)
return jobs
def create_rmap_file(self):
"""
rmap file for Chicago capture analysis is created using the hicup digestion file.
"""
## return 1 rmap per enzyme
output = os.path.join(self.output_dirs['chicago_input_files'],
self.enzyme + ".Initialrmap")
sorted_output = re.sub("\.Initialrmap", ".sorted.rmap", output)
input_file = self.genome_digest
command = """mkdir -p {dir} && \\
cut -f 1-3 {input_file} > {output}.tmp && \\
awk 'BEGIN {{FS=\"\\t\"; OFS=\"\\t\"}} NR>2 {{if ($2 != $3) print $0, NR}}' {output}.tmp > {output} && \\
rm {output}.tmp""".format(dir=self.output_dirs['chicago_input_files'],
input_file=input_file,
output=output)
job_rmap = Job(input_files=[input_file],
output_files=[output, sorted_output],
command=command,
name="create_rmap_file." + self.enzyme)
job_sort = bedops.sort_bed(output, sorted_output)
job = concat_jobs([job_rmap, job_sort])
job.name = "create_rmap_file." + os.path.basename(input_file)
return [job]
def create_baitmap_file(self):
"""
baitmap file for Chicago capture analysis is created using the created rmap file and the probe capture bed file.
"""
## return 1 baitmap per enzyme/capture array combination
input_rmap = os.path.join(self.output_dirs['chicago_input_files'],
self.enzyme + ".sorted.rmap")
input_bait = config.param('create_baitmap_file', "baitBed")
sorted_input_bait = re.sub("\.bed", ".sorted.bed", input_bait)
output_file_name = re.sub(
"\.bed", "",
os.path.basename(input_bait)) + "_" + self.enzyme + ".baitmap"
output_file = os.path.join(self.output_dirs['chicago_input_files'],
output_file_name)
annotation = config.param('create_baitmap_file', "annotation")
job_sort = bedops.sort_bed(input_bait, sorted_input_bait)
job_intersectBeds = bedtools.intersect_beds(input_rmap,
sorted_input_bait,
output_file + ".tmp",
"-wa -u")
## annotate file with annotation in baitBed otherwise annotate with random ids
job_anno = Job(
input_files=[output_file + ".tmp"],
output_files=[output_file],
command="""column_num=$(awk 'NR <2 {{print NF}}' {input_bait})
if [[ \"$column_num\" -eq 4 ]]; then
bedmap --echo --echo-map-id {outputTmp} {sorted_input_bait} > {outputTmp2}
tr '|' '\\t' < {outputTmp2} > {output_file};
rm {outputTmp2};
else
awk 'BEGIN {{FS=\"\\t\"; OFS=\"\\t\"}}{{print $0, \"{annotation}\"NR}}' {outputTmp} > {output_file}
fi""".format(input_bait=input_bait,
outputTmp=output_file + ".tmp",
outputTmp2=output_file + ".tmp2",
sorted_input_bait=sorted_input_bait,
annotation=annotation,
output_file=output_file),
removable_files=[output_file + ".tmp"],
name="create_baitmap_file.addAnno." + output_file_name)
job = concat_jobs([job_sort, job_intersectBeds, job_anno])
job.name = "create_baitmap_file." + output_file_name
return [job]
def create_rmap_file(self):
"""
rmap file for Chicago capture analysis is created using the hicup digestion file.
"""
## return 1 rmap per enzyme
output = os.path.join(self.output_dirs['chicago_input_files'], self.enzyme + ".Initialrmap")
sorted_output = re.sub("\.Initialrmap", ".sorted.rmap", output)
input_file = self.genome_digest
job = concat_jobs([
Job(command="mkdir -p " + self.output_dirs['chicago_input_files']),
tools.sh_create_rmap(input_file, output, "create_rmap_file." + self.enzyme),
bedops.sort_bed(output, sorted_output)
])
job.name = "create_rmap_file." + os.path.basename(input_file)
job.samples = self.samples
return [job]
def create_baitmap_file(self):
"""
baitmap file for Chicago capture analysis is created using the created rmap file and the probe capture bed file.
"""
## return 1 baitmap per enzyme/capture array combination
input_rmap = os.path.join(self.output_dirs['chicago_input_files'], self.enzyme + ".sorted.rmap")
input_bait = config.param('create_baitmap_file', "baitBed")
sorted_input_bait = re.sub("\.bed", ".sorted.bed", input_bait)
output_file_name = re.sub("\.bed", "", os.path.basename(input_bait)) + "_" + self.enzyme + ".baitmap"
output_file = os.path.join(self.output_dirs['chicago_input_files'], output_file_name)
annotation = config.param('create_baitmap_file', "annotation")
job = concat_jobs([
bedops.sort_bed(input_bait, sorted_input_bait),
bedtools.intersect_beds(input_rmap, sorted_input_bait, output_file + ".tmp", "-wa -u"),
tools.sh_create_baitmap(input_bait, sorted_input_bait, annotation, output_file)
])
job.name = "create_baitmap_file." + output_file_name
job.samples = self.samples
return [job]
def capture_intersect(self):
"""
provided with a bed file, for example a bed of GWAS snps or features of interest, this method returns the lines in the bed file that intersect with the captured ends ("Other Ends") that have significant interactions.
Input bed must have 4 columns (<chr> <start> <end> <annotation>) and must be tab separated.
"""
jobs = []
chicago_output_dir = self.output_dirs['chicago_output_directory']
intersect_output_dir = self.output_dirs['intersect_ouput_directory']
other_options = config.param('capture_intersect',
'other_options',
required=False)
features_file = config.param('capture_intersect',
'features_file',
required=True)
sorted_features_file = os.path.splitext(
features_file)[0] + ".sorted.bed"
output_dir = os.path.join(chicago_output_dir, intersect_output_dir)
if features_file != "None":
job_create_dir = Job(command="mkdir -p {output_dir}".format(
output_dir=output_dir))
job_sort_features_file = bedops.sort_bed(features_file,
sorted_features_file)
job = concat_jobs([job_create_dir, job_sort_features_file])
job.name = "capture_intersect.sort_feature." + os.path.basename(
sorted_features_file)
job.removable_files = [sorted_features_file]
jobs.append(job)
for sample in self.samples:
sample_output_dir = os.path.join(chicago_output_dir,
sample.name, "data")
ibed_file = os.path.join(sample_output_dir,
sample.name + ".ibed")
sorted_ibed_file = re.sub("\.ibed$", ".capture.sorted.bed",
ibed_file)
output_file_prefix = os.path.join(
output_dir,
os.path.splitext(os.path.basename(features_file))[0] +
"_" + os.path.splitext(os.path.basename(ibed_file))[0])
job_extract_capture_bed = Job(
input_files=[ibed_file],
output_files=[ibed_file + ".capture"],
name="extract_capture_bed." + sample.name,
command=
"""awk 'BEGIN {{FS=\"\\t\"; OFS=\"\\t\"}} NR>1 {{if ($8 == ".") {{id = $5":"$6"-"$7}} else {{id = $8}} print $5,$6,$7,id}}' {input} > {outputTmp} && \\
awk '!a[$0]++' {outputTmp} > {output} && \\
rm {outputTmp}""".format(
input=ibed_file,
outputTmp=ibed_file + ".capture.tmp",
output=ibed_file + ".capture"),
removable_files=[ibed_file + ".capture"])
job_sort_ibed = bedops.sort_bed(ibed_file + ".capture",
sorted_ibed_file)
job_intersect = bedtools.intersect_beds(
sorted_features_file, sorted_ibed_file,
output_file_prefix + ".tmp", "-wa -u")
job_bedopsMap = bedops.bedmap_echoMapId(
output_file_prefix + ".tmp", sorted_ibed_file,
output_file_prefix + ".capture_intersect.bed")
job = concat_jobs([
job_extract_capture_bed, job_sort_ibed, job_intersect,
job_bedopsMap
])
job.name = "capture_intersect." + sample.name
job.removable_files = [
sorted_ibed_file, output_file_prefix + ".tmp",
ibed_file + ".capture"
]
jobs.append(job)
return jobs