def bed_to_bg(input_file, output_file, size_file, extras):
log.info("Converting %s to a genome coverage file", input_file)
args = ["genomeCoverageBed", "-bg", "-split", "-i", input_file, "-g", size_file, ">", output_file]
utils.run_cmd(args, "bed_to_bg")
log.info("Deleting old file %s", input_file)
os.unlink(input_file)
def pear_fastq_files(input_files, output_file, extras):
log.info("Starting pear run on %s and %s", input_files[0], input_files[1])
output_file = re.sub(r"\.assembled\.fastq", "", output_file)
args = ["pear", "-f", input_files[0], "-r", input_files[1], "-o", output_file]
utils.run_cmd("pear")
def bam_to_bed(input_file, output_file, output, extras):
log.info("Converting %s to a bed file", input_file)
args = ["bamToBed", "-i", input_file ">", output_file]
utils.run_cmd(args, "bam_to_bed")
# now we can move sorted bam to output
file_name = os.path.basename(input_file)
new_name = os.path.join(output, file_name)
os.rename(input_file, new_name)
def bw_stats(input_files, output_file):
# we are going to call bwtool summary and bwtool distribution
# have to explicitly send stdout stuff like that
# what a program
summary = "bwtool summary 10000 -header -with-sum {} /dev/stdout"
dist = "bwtool distribution {} /dev/stdout"
for input_file in input_files:
log.info("Running bigwig stats on {}".format(input_file))
with open(output_file, "a+") as stats:
for command in [summary, dist]:
output = utils.run_cmd(command.format(os.path.abspath(input_file)).split(), command, True)
if command.startswith("bwtool summary"):
stats.write("#### bwtool summary for {}\n".format(input_file))
stats.write(output)
stats.write("####\n")
# filter zeros out
else:
output = output.rstrip()
output_clean = [line for line in output.split("\n") if line.split('\t')[1] != '0']
stats.write("#### bwtool distribution for {}\n".format(input_file))
stats.write("depth\tcount\n")
stats.write("\n".join(output_clean))
stats.write("\n####\n")
stats.write("\n\n")
def align_with_bowtie_two(input_file, output_file, options, stats_file, extras):
log.info("Running bowtie2 on %s", input_file)
# use poen explicitly to cpature STDERR, check still
args = ["bowtie2", "-t", "--no-unal", "-p", str(options.cores), "-x", options.index, input_file, "-S", output_file]
output = utils.run_cmd(args, "bowtie2", True)
# pass along to be saved
utils.bowtie_record_output(output, input_file, stats_file)
def upload_to_one_codex(input_file, output_file, extras):
args = ["onecodex", "upload", input_file]
log.info("uploading %s to One Codex", input_file)
utils.run_cmd(args, "One Codex")
args = ["onecodex", "upload", input_file]
log.info("uploading %s to One Codex", input_file)
utils.run_cmd(args, "One Codex")
# only bowtie
# rsem does this for you
@transform(rename_accepted_hits, suffix(".bam"),".sorted.bam", extras)
def sort_bam(input_file, output_file, extras):
log.info("Sorting %s ", input_file)
# hacky
output_file = re.sub(r"\.bam", "", output_file)
args = ["samtools-rs", "rocksort", "-@", "8", "-m", "16G", input_file, output_file]):
utils.run_cmd(args, "samtools rocksort")
# careful
log.info("Deleting old file %s", input_file)
os.unlink(input_file)
@transform(sort_bam, suffix(".sorted.bam"), ".bed", options.output, extras)
def bam_to_bed(input_file, output_file, output, extras):
log.info("Converting %s to a bed file", input_file)
args = ["bamToBed", "-i", input_file ">", output_file]
utils.run_cmd(args, "bam_to_bed")
# now we can move sorted bam to output
file_name = os.path.basename(input_file)