def work(self):
"""
Worker function for filtering bam files using BioBamBam
Parameters
----------
in_bam_file : str
Location of the bam file to filter
in_bam_file : str
Location of the filtered bam file
"""
macs2_handle = macs2()
real_params = []
if len(self.macs2_params) > 0:
real_params = self.macs2_params.split(",")
macs2_handle.macs2_peak_calling_nobgd(
"luigi_lsf", self.in_bam_file, real_params, self.narrowPeak_file,
self.summits_file, self.broadPeak_file, self.gappedPeak_file)
def test_macs2_background():
"""
Function to test MACS2
"""
resource_path = os.path.join(os.path.dirname(__file__), "data/")
input_files = {
"bam": resource_path + "macs2.Human.DRR000150.22_aln_filtered.bam",
"bam_bg": resource_path + "macs2.Human.DRR000150.22_aln_filtered.bam"
}
output_files = {
"narrow_peak":
resource_path + "macs2.Human.DRR000150.22_peaks.narrowPeak",
"summits":
resource_path + "macs2.Human.DRR000150.22_peaks.summits.bed",
"broad_peak":
resource_path + "macs2.Human.DRR000150.22_peaks.broadPeak",
"gapped_peak":
resource_path + "macs2.Human.DRR000150.22_peaks.gappedPeak"
}
metadata = {
"bam":
Metadata("data_chipseq", "fastq", [], None, {'assembly': 'test'}),
"bam_bg":
Metadata("data_chipseq", "fastq", [], None, {'assembly': 'test'}),
}
macs_handle = macs2({"macs_nomodel_param": True})
macs_handle.run(input_files, metadata, output_files)
assert os.path.isfile(resource_path +
"macs2.Human.DRR000150.22_peaks.narrowPeak") is True
assert os.path.getsize(resource_path +
"macs2.Human.DRR000150.22_peaks.narrowPeak") > 0
assert os.path.isfile(resource_path +
"macs2.Human.DRR000150.22_peaks.summits.bed") is True
assert os.path.getsize(resource_path +
"macs2.Human.DRR000150.22_peaks.summits.bed") > 0
def run(self, input_files, metadata, output_files):
"""
Main run function for processing ChIP-seq FastQ data. Pipeline aligns
the FASTQ files to the genome using BWA. MACS 2 is then used for peak
calling to identify transcription factor binding sites within the
genome.
Currently this can only handle a single data file and a single
background file.
Parameters
----------
input_files : dict
Location of the initial input files required by the workflow
bam : str
Location of the aligned reads file
bam_bg : str
Location of the background aligned FASTQ reads file [OPTIONAL]
metadata : dict
Input file meta data associated with their roles
bam : str
bam_bg : str
[OPTIONAL]
output_files : dict
Output file locations
narrow_peak : str
summits : str
broad_peak : str
gapped_peak : str
Returns
-------
output_files : dict
Output file locations associated with their roles, for the output
narrow_peak : str
Results files in bed4+1 format
summits : str
Results files in bed6+4 format
broad_peak : str
Results files in bed6+3 format
gapped_peak : str
Results files in bed12+3 format
output_metadata : dict
Output metadata for the associated files in output_files
narrow_peak : Metadata
summits : Metadata
broad_peak : Metadata
gapped_peak : Metadata
"""
output_files_generated = {}
output_metadata = {}
# MACS2 to call peaks
macs_caller = macs2(self.configuration)
macs_inputs = {"bam": input_files["bam"]}
macs_metadt = {"bam": metadata['bam']}
if "bg_loc" in input_files:
macs_inputs["bam"] = input_files["bam_bg"]
macs_metadt["bam"] = output_metadata['bam_bg']
m_results_files, m_results_meta = macs_caller.run(
macs_inputs,
macs_metadt,
# Outputs of the final step may match workflow outputs;
# Extra entries in output_files will be disregarded.
remap(output_files, 'narrow_peak', 'summits', 'broad_peak',
'gapped_peak'))
if 'narrow_peak' in m_results_meta:
output_files_generated['narrow_peak'] = m_results_files[
'narrow_peak']
output_metadata['narrow_peak'] = m_results_meta['narrow_peak']
tool_name = output_metadata['narrow_peak'].meta_data['tool']
output_metadata['narrow_peak'].meta_data[
'tool_description'] = tool_name
output_metadata['narrow_peak'].meta_data[
'tool'] = "process_chipseq"
if 'summits' in m_results_meta:
output_files_generated['summits'] = m_results_files['summits']
output_metadata['summits'] = m_results_meta['summits']
tool_name = output_metadata['summits'].meta_data['tool']
output_metadata['summits'].meta_data[
'tool_description'] = tool_name
output_metadata['summits'].meta_data['tool'] = "process_chipseq"
if 'broad_peak' in m_results_meta:
output_files_generated['broad_peak'] = m_results_files[
'broad_peak']
output_metadata['broad_peak'] = m_results_meta['broad_peak']
tool_name = output_metadata['broad_peak'].meta_data['tool']
output_metadata['broad_peak'].meta_data[
'tool_description'] = tool_name
output_metadata['broad_peak'].meta_data['tool'] = "process_chipseq"
if 'gapped_peak' in m_results_meta:
output_files_generated['gapped_peak'] = m_results_files[
'gapped_peak']
output_metadata['gapped_peak'] = m_results_meta['gapped_peak']
tool_name = output_metadata['gapped_peak'].meta_data['tool']
output_metadata['gapped_peak'].meta_data[
'tool_description'] = tool_name
output_metadata['gapped_peak'].meta_data[
'tool'] = "process_chipseq"
return output_files_generated, output_metadata
def run(self, input_files, metadata, output_files): # pylint: disable=too-many-branches
"""
Main run function for processing ChIP-seq FastQ data. Pipeline aligns
the FASTQ files to the genome using BWA. MACS 2 is then used for peak
calling to identify transcription factor binding sites within the
genome.
Currently this can only handle a single data file and a single
background file.
Parameters
----------
input_files : dict
Location of the initial input files required by the workflow
genome : str
Genome FASTA file
index : str
Location of the BWA archived index files
loc : str
Location of the FASTQ reads files
fastq2 : str
Location of the paired end FASTQ file [OPTIONAL]
bg_loc : str
Location of the background FASTQ reads files [OPTIONAL]
fastq2_bg : str
Location of the paired end background FASTQ reads files [OPTIONAL]
metadata : dict
Input file meta data associated with their roles
genome : str
index : str
bg_loc : str
[OPTIONAL]
output_files : dict
Output file locations
bam [, "bam_bg"] : str
filtered [, "filtered_bg"] : str
narrow_peak : str
summits : str
broad_peak : str
gapped_peak : str
Returns
-------
output_files : dict
Output file locations associated with their roles, for the output
bam [, "bam_bg"] : str
Aligned FASTQ short read file [ and aligned background file]
locations
filtered [, "filtered_bg"] : str
Filtered versions of the respective bam files
narrow_peak : str
Results files in bed4+1 format
summits : str
Results files in bed6+4 format
broad_peak : str
Results files in bed6+3 format
gapped_peak : str
Results files in bed12+3 format
output_metadata : dict
Output metadata for the associated files in output_files
bam [, "bam_bg"] : Metadata
filtered [, "filtered_bg"] : Metadata
narrow_peak : Metadata
summits : Metadata
broad_peak : Metadata
gapped_peak : Metadata
"""
output_files_generated = {}
output_metadata = {}
logger.info("PROCESS CHIPSEQ - DEFINED OUTPUT:", output_files["bam"])
align_input_files = remap(input_files, "genome", "loc", "index")
align_input_file_meta = remap(metadata, "genome", "loc", "index")
if "fastq2" in input_files:
align_input_files["fastq2"] = input_files["fastq2"]
align_input_file_meta["fastq2"] = metadata["fastq2"]
bwa = bwaAlignerTool(self.configuration)
bwa_files, bwa_meta = bwa.run(align_input_files, align_input_file_meta,
{"output": output_files["bam"]})
try:
output_files_generated["bam"] = bwa_files["bam"]
output_metadata["bam"] = bwa_meta["bam"]
tool_name = output_metadata['bam'].meta_data['tool']
output_metadata['bam'].meta_data['tool_description'] = tool_name
output_metadata['bam'].meta_data['tool'] = "process_chipseq"
except KeyError:
logger.fatal("BWA aligner failed")
if "bg_loc" in input_files:
# Align background files
align_input_files_bg = remap(input_files,
"genome",
"index",
loc="bg_loc")
align_input_file_meta_bg = remap(metadata,
"genome",
"index",
loc="bg_loc")
if "fastq2" in input_files:
align_input_files_bg["fastq2"] = input_files["fastq2_bg"]
align_input_file_meta_bg["fastq2"] = metadata["fastq2_bg"]
bwa_bg_files, bwa_bg_meta = bwa.run(
align_input_files_bg, align_input_file_meta_bg,
{"output": output_files["bam_bg"]})
try:
output_files_generated["bam_bg"] = bwa_bg_files["bam_bg"]
output_metadata["bam_bg"] = bwa_bg_meta["bam_bg"]
tool_name = output_metadata['bam_bg'].meta_data['tool']
output_metadata['bam_bg'].meta_data[
'tool_description'] = tool_name
output_metadata['bam_bg'].meta_data['tool'] = "process_chipseq"
except KeyError:
logger.fatal("Background BWA aligner failed")
# Filter the bams
b3f = biobambam(self.configuration)
b3f_files, b3f_meta = b3f.run({"input": bwa_files['bam']},
{"input": bwa_meta['bam']},
{"output": output_files["filtered"]})
try:
output_files_generated["filtered"] = b3f_files["bam"]
output_metadata["filtered"] = b3f_meta["bam"]
tool_name = output_metadata['filtered'].meta_data['tool']
output_metadata['filtered'].meta_data[
'tool_description'] = tool_name
output_metadata['filtered'].meta_data['tool'] = "process_chipseq"
except KeyError:
logger.fatal("BioBamBam filtering failed")
if "bg_loc" in input_files:
# Filter background aligned files
b3f_bg_files, b3f_bg_meta = b3f.run(
{"input": bwa_bg_files['bam']}, {"input": bwa_bg_meta['bam']},
{"output": output_files["filtered_bg"]})
try:
output_files_generated["filtered_bg"] = b3f_bg_files["bam"]
output_metadata["filtered_bg"] = b3f_bg_meta["bam"]
tool_name = output_metadata['filtered_bg'].meta_data['tool']
output_metadata['filtered_bg'].meta_data[
'tool_description'] = tool_name
output_metadata['filtered_bg'].meta_data[
'tool'] = "process_chipseq"
except KeyError:
logger.fatal("Background BioBamBam filtering failed")
# MACS2 to call peaks
macs_caller = macs2(self.configuration)
macs_inputs = {"bam": output_files_generated["filtered"]}
macs_metadt = {"bam": output_metadata['filtered']}
if "bg_loc" in input_files:
macs_inputs["bam_bg"] = output_files_generated["filtered_bg"]
macs_metadt["bam_bg"] = output_metadata['filtered_bg']
m_results_files, m_results_meta = macs_caller.run(
macs_inputs,
macs_metadt,
# Outputs of the final step may match workflow outputs;
# Extra entries in output_files will be disregarded.
remap(output_files, 'narrow_peak', 'summits', 'broad_peak',
'gapped_peak'))
if not m_results_meta:
logger.fatal("MACS2 peak calling failed")
if 'narrow_peak' in m_results_meta:
output_files_generated['narrow_peak'] = m_results_files[
'narrow_peak']
output_metadata['narrow_peak'] = m_results_meta['narrow_peak']
tool_name = output_metadata['narrow_peak'].meta_data['tool']
output_metadata['narrow_peak'].meta_data[
'tool_description'] = tool_name
output_metadata['narrow_peak'].meta_data[
'tool'] = "process_chipseq"
if 'summits' in m_results_meta:
output_files_generated['summits'] = m_results_files['summits']
output_metadata['summits'] = m_results_meta['summits']
tool_name = output_metadata['summits'].meta_data['tool']
output_metadata['summits'].meta_data[
'tool_description'] = tool_name
output_metadata['summits'].meta_data['tool'] = "process_chipseq"
if 'broad_peak' in m_results_meta:
output_files_generated['broad_peak'] = m_results_files[
'broad_peak']
output_metadata['broad_peak'] = m_results_meta['broad_peak']
tool_name = output_metadata['broad_peak'].meta_data['tool']
output_metadata['broad_peak'].meta_data[
'tool_description'] = tool_name
output_metadata['broad_peak'].meta_data['tool'] = "process_chipseq"
if 'gapped_peak' in m_results_meta:
output_files_generated['gapped_peak'] = m_results_files[
'gapped_peak']
output_metadata['gapped_peak'] = m_results_meta['gapped_peak']
tool_name = output_metadata['gapped_peak'].meta_data['tool']
output_metadata['gapped_peak'].meta_data[
'tool_description'] = tool_name
output_metadata['gapped_peak'].meta_data[
'tool'] = "process_chipseq"
return output_files_generated, output_metadata