def genebody_coverage2(in_file, config, out_prefix=None):
"""
used to check the 5'/3' bias across transcripts, takes a bam file,
converts it to bigwig and then uses that
"""
PROGRAM = "geneBody_coverage2.py"
if not program_exists(PROGRAM):
logger.info("%s is not in the path or is not executable." % (PROGRAM))
exit(1)
in_bigwig = bam2bigwig(in_file, config)
prefix = "coverage"
out_dir = os.path.join(os.path.dirname(in_bigwig), os.pardir, "coverage")
safe_makedir(out_dir)
out_prefix = out_dir + "/wiggle"
#out_prefix = _get_out_prefix(in_bigwig, config, out_prefix, prefix)
coverage_plot_file = out_prefix + ".geneBodyCoverage.pdf"
if file_exists(coverage_plot_file):
return coverage_plot_file
gtf = _get_gtf(config)
bed = _gtf2bed(gtf)
coverage_run = sh.Command(which(PROGRAM))
cmd = str(coverage_run.bake(i=in_bigwig, r=bed, o=out_prefix, t="pdf"))
do.run(cmd, "Calculating coverage of %s." % (in_bigwig), None)
return coverage_plot_file
def _build_command(input_file, ref, novoalign_config):
cmd = [
which("novoalign"),
flatten_options(novoalign_config), "-o", "SAM", "-d", ref, "-f",
input_file
]
return list(map(str, flatten(cmd)))
def wig2bigwig(wiggle_file, chrom_size_file, out_file):
"""
convert wiggle file to bigwig file using the UCSC tool
"""
PROGRAM = "wigToBigWig"
if not program_exists(PROGRAM):
logger.error("%s is not in the path or is not executable. Make sure "
"it is installed or go to "
"http://hgdownload.cse.ucsc.edu/admin/exe/"
"to download it." % (PROGRAM))
exit(1)
if file_exists(out_file):
return out_file
wigToBigWig = sh.Command(which(PROGRAM))
with file_transaction(out_file) as tx_out_file:
wigToBigWig(wiggle_file, chrom_size_file, tx_out_file)
return out_file
def junction_annotation(in_file, config, out_prefix=None):
"""
compile novel/known information about splice junctions
"""
PROGRAM = "junction_annotation.py"
if not program_exists(PROGRAM):
logger.info("%s is not in the path or is not executable." % (PROGRAM))
exit(1)
prefix = "junction"
out_prefix = _get_out_prefix(in_file, config, out_prefix, prefix)
junction_file = out_prefix + ".splice_junction.pdf"
if file_exists(junction_file):
return junction_file
junction_run = sh.Command(which(PROGRAM))
gtf = _get_gtf(config)
bed = _gtf2bed(gtf)
junction_run(i=in_file, o=out_prefix, r=bed)
return junction_file
def RPKM_count(in_file, config, out_prefix=None):
"""
produce RPKM
"""
PROGRAM = "RPKM_count.py"
if not program_exists(PROGRAM):
logger.info("%s is not in the path or is not executable." % (PROGRAM))
exit(1)
prefix = "RPKM_count"
out_prefix = _get_out_prefix(in_file, config, out_prefix, prefix)
rpkm_count_file = out_prefix + "_read_count.xls"
gtf = _get_gtf(config)
bed = _gtf2bed(gtf)
if file_exists(rpkm_count_file):
return rpkm_count_file
RPKM_count_run = sh.Command(which(PROGRAM))
RPKM_count_run(i=in_file, r=bed, o=out_prefix)
return rpkm_count_file
def genebody_coverage(in_file, config, out_prefix=None):
"""
used to check the 5'/3' bias across transcripts
"""
PROGRAM = "geneBody_coverage.py"
if not program_exists(PROGRAM):
logger.info("%s is not in the path or is not executable." % (PROGRAM))
exit(1)
prefix = "coverage"
out_prefix = _get_out_prefix(in_file, config, out_prefix, prefix)
coverage_plot_file = out_prefix + ".geneBodyCoverage.pdf"
if file_exists(coverage_plot_file):
return coverage_plot_file
gtf = _get_gtf(config)
bed = _gtf2bed(gtf)
coverage_run = sh.Command(which(PROGRAM))
coverage_run(i=in_file, r=bed, o=out_prefix)
return coverage_plot_file
def bam_stat(in_file, config, out_prefix=None):
"""
dump read maping statistics from a SAM or BAM file to out_file
"""
PROGRAM = "bam_stat.py"
if not program_exists(PROGRAM):
logger.info("%s is not in the path or is not executable." % (PROGRAM))
exit(1)
prefix = "bam_stat"
out_prefix = _get_out_prefix(in_file, config, out_prefix, prefix)
out_file = out_prefix + ".txt"
if file_exists(out_file):
return out_file
bam_stat_run = sh.Command(which(PROGRAM))
with file_transaction(out_file) as tx_out_file:
bam_stat_run(i=in_file, _err=tx_out_file)
return out_file
def wig2bigwig(wiggle_file, chrom_size_file, out_file):
"""
convert wiggle file to bigwig file using the UCSC tool
"""
PROGRAM = "wigToBigWig"
if not program_exists(PROGRAM):
logger.error("%s is not in the path or is not executable. Make sure "
"it is installed or go to "
"http://hgdownload.cse.ucsc.edu/admin/exe/"
"to download it." % (PROGRAM))
exit(1)
if file_exists(out_file):
return out_file
wigToBigWig = sh.Command(which(PROGRAM))
with file_transaction(out_file) as tx_out_file:
cmd = str(wigToBigWig.bake(wiggle_file, chrom_size_file, tx_out_file))
do.run(cmd, "Converting %s from wig to bigwig." % (wiggle_file), None)
return out_file
def count_overlaps(in_file, bed, out_file=None):
""" calculates coverage across the features in the bedfile
bed """
if not which("coverageBed"):
logger.error("Cannot find coverageBed. Make sure it is in your "
"path or install bedtools.")
exit(-1)
if not out_file:
out_file = replace_suffix(in_file, ".counts")
if os.path.exists(out_file):
return out_file
cmd = ["coverageBed", "-abam", in_file, "-b", bed]
with open(out_file, "w") as out_handle:
subprocess.check_call(cmd, stdout=out_handle)
return out_file
def count_overlaps(in_file, bed, out_file=None):
""" calculates coverage across the features in the bedfile
bed """
if not which("coverageBed"):
logger.error("Cannot find coverageBed. Make sure it is in your "
"path or install bedtools.")
exit(-1)
if not out_file:
out_file = replace_suffix(in_file, ".counts")
if os.path.exists(out_file):
return out_file
cmd = ["coverageBed", "-abam", in_file, "-b", bed]
with open(out_file, "w") as out_handle:
subprocess.check_call(cmd, stdout=out_handle)
return out_file
def junction_saturation(in_file, config, out_prefix=None):
"""
check if splicing is deep enough to perform alternative splicing
analysis
"""
PROGRAM = "junction_saturation.py"
if not program_exists(PROGRAM):
logger.info("%s is not in the path or is not executable." % (PROGRAM))
exit(1)
prefix = "saturation"
out_prefix = _get_out_prefix(in_file, config, out_prefix, prefix)
saturation_file = out_prefix + ".junctionSaturation_plot.pdf"
if file_exists(saturation_file):
return saturation_file
saturation_run = sh.Command(which(PROGRAM))
gtf = _get_gtf(config)
bed = _gtf2bed(gtf)
saturation_run(i=in_file, o=out_prefix, r=bed)
return saturation_file
def genebody_coverage(in_file, config, out_prefix=None):
"""
used to check the 5'/3' bias across transcripts
"""
PROGRAM = "geneBody_coverage.py"
if not program_exists(PROGRAM):
logger.info("%s is not in the path or is not executable." % (PROGRAM))
exit(1)
prefix = "coverage"
out_prefix = _get_out_prefix(in_file, config, out_prefix, prefix)
coverage_plot_file = out_prefix + ".geneBodyCoverage.pdf"
if file_exists(coverage_plot_file):
return coverage_plot_file
gtf = _get_gtf(config)
bed = _gtf2bed(gtf)
coverage_run = sh.Command(which(PROGRAM))
cmd = str(coverage_run.bake(i=in_file, r=bed, o=out_prefix))
do.run(cmd, "Calculating coverage of %s." % (in_file), None)
return coverage_plot_file
def RPKM_saturation(in_file, config, out_prefix=None):
"""
estimate the precision of RPKM calculation by resampling
"""
PROGRAM = "RPKM_saturation.py"
if not program_exists(PROGRAM):
logger.info("%s is not in the path or is not executable." % (PROGRAM))
exit(1)
prefix = "RPKM_saturation"
out_prefix = _get_out_prefix(in_file, config, out_prefix, prefix)
rpkm_saturation_file = out_prefix + ".saturation.pdf"
gtf = _get_gtf(config)
bed = _gtf2bed(gtf)
if file_exists(rpkm_saturation_file):
return rpkm_saturation_file
RPKM_saturation_run = sh.Command(which(PROGRAM))
RPKM_saturation_run(i=in_file, r=bed, o=out_prefix)
return rpkm_saturation_file
def clipping_profile(in_file, config, out_prefix=None):
"""
estimate the clipping profile of the reads
"""
PROGRAM = "clipping_profile.py"
if not program_exists(PROGRAM):
logger.info("%s is not in the path or is not executable." % (PROGRAM))
exit(1)
prefix = "clipping"
out_prefix = _get_out_prefix(in_file, config, out_prefix, "clipping")
clip_plot_file = out_prefix + ".clipping_profile.pdf"
print clip_plot_file
if file_exists(clip_plot_file):
return clip_plot_file
clip_run = sh.Command(which(PROGRAM))
cmd = str(clip_run.bake(i=in_file, o=out_prefix))
do.run(cmd, "Calculating 5' and 3' clipping profile of %s." % (in_file), None)
return clip_plot_file
def RPKM_count(in_file, config, out_prefix=None):
"""
produce RPKM
"""
PROGRAM = "RPKM_count.py"
if not program_exists(PROGRAM):
logger.info("%s is not in the path or is not executable." % (PROGRAM))
exit(1)
prefix = "RPKM_count"
out_prefix = _get_out_prefix(in_file, config, out_prefix, prefix)
rpkm_count_file = out_prefix + "_read_count.xls"
gtf = _get_gtf(config)
bed = _gtf2bed(gtf)
if file_exists(rpkm_count_file):
return rpkm_count_file
RPKM_count_run = sh.Command(which(PROGRAM))
cmd = str(RPKM_count_run.bake(i=in_file, r=bed, o=out_prefix))
do.run(cmd, "Calculating RPKM of %s using reference %s." % (in_file, bed),
None)
return rpkm_count_file
def junction_annotation(in_file, config, out_prefix=None):
"""
compile novel/known information about splice junctions
"""
PROGRAM = "junction_annotation.py"
if not program_exists(PROGRAM):
logger.info("%s is not in the path or is not executable." % (PROGRAM))
exit(1)
prefix = "junction"
out_prefix = _get_out_prefix(in_file, config, out_prefix, prefix)
junction_file = out_prefix + ".splice_junction.pdf"
if file_exists(junction_file):
return junction_file
junction_run = sh.Command(which(PROGRAM))
gtf = _get_gtf(config)
bed = _gtf2bed(gtf)
cmd = str(junction_run.bake(i=in_file, o=out_prefix, r=bed))
do.run(cmd, "Calculating novel/known information about splice junctions of "
"%s." % (in_file), None)
return junction_file
def bam_stat(in_file, config, out_prefix=None):
"""
dump read maping statistics from a SAM or BAM file to out_file
"""
PROGRAM = "bam_stat.py"
if not program_exists(PROGRAM):
logger.info("%s is not in the path or is not executable." % (PROGRAM))
exit(1)
prefix = "bam_stat"
out_prefix = _get_out_prefix(in_file, config, out_prefix, prefix)
out_file = out_prefix + ".txt"
if file_exists(out_file):
return out_file
bam_stat_run = sh.Command(which(PROGRAM))
with file_transaction(out_file) as tx_out_file:
cmd = str(bam_stat_run.bake(i=in_file, _err=tx_out_file))
do.run(cmd, "Calculating BAM statistics from %s." % (in_file), None)
return out_file
def RPKM_saturation(in_file, config, out_prefix=None):
"""
estimate the precision of RPKM calculation by resampling
"""
PROGRAM = "RPKM_saturation.py"
if not program_exists(PROGRAM):
logger.info("%s is not in the path or is not executable." % (PROGRAM))
exit(1)
prefix = "RPKM_saturation"
out_prefix = _get_out_prefix(in_file, config, out_prefix, prefix)
rpkm_saturation_file = out_prefix + ".saturation.pdf"
gtf = _get_gtf(config)
bed = _gtf2bed(gtf)
if file_exists(rpkm_saturation_file):
return rpkm_saturation_file
RPKM_saturation_run = sh.Command(which(PROGRAM))
cmd = str(RPKM_saturation_run.bake(i=in_file, r=bed, o=out_prefix))
do.run(cmd, "Calculating RPKM saturation of %s." % in_file, None)
return rpkm_saturation_file
def check_program(self, config):
need_dirs = ["picard"]
programs = config.get("program", None)
if not programs:
self._field_missing_warning("program")
return True
for program, loc in programs.items():
if program in need_dirs:
if os.path.isdir(loc):
logger.info("%s lives in %s." % (program, loc))
else:
self._directory_missing_error(program, loc)
return False
else:
if which(loc):
logger.info("%s will be run as %s." % (program, loc))
else:
self._executable_missing_error(program, loc)
return False
logger.info("The 'program' section of configuration seems valid.")
return True
def junction_saturation(in_file, config, out_prefix=None):
"""
check if splicing is deep enough to perform alternative splicing
analysis
"""
PROGRAM = "junction_saturation.py"
if not program_exists(PROGRAM):
logger.info("%s is not in the path or is not executable." % (PROGRAM))
exit(1)
prefix = "saturation"
out_prefix = _get_out_prefix(in_file, config, out_prefix, prefix)
saturation_file = out_prefix + ".junctionSaturation_plot.pdf"
if file_exists(saturation_file):
return saturation_file
saturation_run = sh.Command(which(PROGRAM))
gtf = _get_gtf(config)
bed = _gtf2bed(gtf)
cmd = str(saturation_run.bake(i=in_file, o=out_prefix, r=bed))
do.run(cmd, "Calculating junction saturation estimation of %s." % in_file,
None)
return saturation_file
def clipping_profile(in_file, config, out_prefix=None):
"""
estimate the clipping profile of the reads
"""
PROGRAM = "clipping_profile.py"
if not program_exists(PROGRAM):
logger.info("%s is not in the path or is not executable." % (PROGRAM))
exit(1)
prefix = "clipping"
out_prefix = _get_out_prefix(in_file, config, out_prefix, "clipping")
clip_plot_file = out_prefix + ".clipping_profile.pdf"
print clip_plot_file
if file_exists(clip_plot_file):
return clip_plot_file
clip_run = sh.Command(which(PROGRAM))
clip_run(i=in_file, o=out_prefix)
# hack to get around the fact that clipping_profile saves the file in
# the script execution directory
#sh.mv("clipping_profile.pdf", clip_plot_file)
return clip_plot_file
def bam2bigwig(in_file, config, out_prefix=None):
"""
assumes the library preparation was not strand specific for now
"""
PROGRAM = "bam2wig.py"
if not program_exists(PROGRAM):
logger.info("%s is not in the path or is not executable." % (PROGRAM))
exit(1)
prefix = "bigwig"
chrom_size_file = config["annotation"].get("chrom_size_file", None)
out_prefix = _get_out_prefix(in_file, config, out_prefix, prefix)
if not chrom_size_file:
chrom_size_file = _fetch_chrom_sizes(config)
wiggle_file = out_prefix + ".wig"
if not file_exists(wiggle_file):
bam2wig = sh.Command(which(PROGRAM))
bam2wig(i=in_file, s=chrom_size_file, o=out_prefix)
bigwig_file = out_prefix + ".bw"
return wig2bigwig(wiggle_file, chrom_size_file, bigwig_file)
def bam2bigwig(in_file, config, out_prefix=None):
"""
assumes the library preparation was not strand specific for now
"""
PROGRAM = "bam2wig.py"
if not program_exists(PROGRAM):
logger.info("%s is not in the path or is not executable." % (PROGRAM))
exit(1)
prefix = "bigwig"
chrom_size_file = config["annotation"].get("chrom_size_file", None)
out_prefix = _get_out_prefix(in_file, config, out_prefix, prefix)
if not chrom_size_file:
chrom_size_file = _fetch_chrom_sizes(config)
wiggle_file = out_prefix + ".wig"
if not file_exists(wiggle_file):
bam2wig = sh.Command(which(PROGRAM))
cmd = str(bam2wig.bake(i=in_file, s=chrom_size_file, o=out_prefix))
do.run(cmd, "Converting %s from BAM to bigwig" % (in_file), None)
bigwig_file = out_prefix + ".bw"
return wig2bigwig(wiggle_file, chrom_size_file, bigwig_file)
def program_exists(path):
return which(path)
def _build_command(input_file, ref, novoalign_config):
cmd = [which("novoalign"), flatten_options(novoalign_config),
"-o", "SAM", "-d", ref, "-f", input_file]
return list(map(str, flatten(cmd)))