def run_bowtie2(paired_end_mapping,
genome,
output_path,
disable_parallel=False):
bowtie2_logger = _logshim.getLogger('run_bowtie2')
# Import the config file to get genome locations
config = _script_helpers.get_config()
if disable_parallel:
shell_job_runner = _script_helpers.ShellJobRunner(bowtie2_logger)
else:
shell_job_runner = _script_helpers.ShellJobRunner(bowtie2_logger,
delay_seconds=60)
for output_prefix, paired_ends in paired_end_mapping.iteritems():
bowtie2_logger.info('Spawning niced process for bowtie2 on: %s' %
(output_prefix))
for filename in paired_ends:
assert (" " not in filename)
assert (";" not in filename
) # Vague sanity testing for input filenames
bowtie2_logger.debug(' Input: %s' % (filename))
# bowtie2 options:
# --end-to-end: this is the default, but let's explicitly specify it
# --sensitive: again, the default (consider switching to --fast?)
# --no-unal: Suppress unaligned reads from the output .sam
# --no-discordant: These are paired-end reads. We expect them to be non-discordant.
# --mm: mmap MAP_SHARED (other processes can use our genome, cool!)
# --met-stderr: Write metrics to stderr
# --time: output the time things took
# -x: target genome
command = "bowtie2 --end-to-end --sensitive --no-unal --no-discordant --mm --met-stderr --time -x %s -1 %s -2 %s 2>%s | samtools view -bS - >%s"
shell_job_runner.run(
command %
(config['bowtie2_genomes'][genome], paired_ends[0], paired_ends[1],
output_path + "/" + output_prefix + ".bt2.log",
output_path + "/" + output_prefix + ".bt2.bam"))
shell_job_runner.finish()
def run_bowtie2(paired_end_mapping, genome, output_path, disable_parallel=False):
bowtie2_logger = _logshim.getLogger('run_bowtie2')
# Import the config file to get genome locations
config = _script_helpers.get_config()
if disable_parallel:
shell_job_runner = _script_helpers.ShellJobRunner(bowtie2_logger)
else:
shell_job_runner = _script_helpers.ShellJobRunner(bowtie2_logger, delay_seconds=60)
for output_prefix, paired_ends in paired_end_mapping.iteritems():
bowtie2_logger.info('Spawning niced process for bowtie2 on: %s' % (output_prefix))
for filename in paired_ends:
assert(" " not in filename)
assert(";" not in filename) # Vague sanity testing for input filenames
bowtie2_logger.debug(' Input: %s' % (filename))
# bowtie2 options:
# --end-to-end: this is the default, but let's explicitly specify it
# --sensitive: again, the default (consider switching to --fast?)
# --no-unal: Suppress unaligned reads from the output .sam
# --no-discordant: These are paired-end reads. We expect them to be non-discordant.
# --mm: mmap MAP_SHARED (other processes can use our genome, cool!)
# --met-stderr: Write metrics to stderr
# --time: output the time things took
# -x: target genome
command = "bowtie2 --end-to-end --sensitive --no-unal --no-discordant --mm --met-stderr --time -x %s -1 %s -2 %s 2>%s | samtools view -bS - >%s"
shell_job_runner.run(command % (config['bowtie2_genomes'][genome],
paired_ends[0],
paired_ends[1],
output_path + "/" + output_prefix + ".bt2.log",
output_path + "/" + output_prefix + ".bt2.bam"))
shell_job_runner.finish()
__copyright__ = 'Gordon Lab at Washington University in St. Louis'
__license__ = 'MIT'
__version__ = '1.0.3'
import _logshim
import _script_helpers
import argparse
import glob
import os
import tempfile
# A parameter needed by samtools to sort in-memory.
MAX_MEM = "50G"
# Load our config files
CONFIG = _script_helpers.get_config()
def large_filter_fixmate_and_sort(input_files, genome, output_path, disable_parallel=False):
primary_logger = _logshim.getLogger('first_pass')
output_suffix = ".tmp"
if disable_parallel: # Doesn't change parallelism in last samtools sort
shell_job_runner = _script_helpers.ShellJobRunner(primary_logger)
else:
shell_job_runner = _script_helpers.ShellJobRunner(primary_logger, delay_seconds=60)
# We do a few things here:
# - View only mapping quality >= 10
# - Remove chrM