def main_deplete_human(args):
''' Run the entire depletion pipeline: bmtagger, mvicuna, blastn.
Optionally, use lastal to select a specific taxon of interest.'''
tools.picard.RevertSamTool().execute(
args.inBam,
args.revertBam,
picardOptions=['SORT_ORDER=queryname', 'SANITIZE=true'])
multi_db_deplete_bam(args.revertBam,
args.bmtaggerDbs,
deplete_bmtagger_bam,
args.bmtaggerBam,
threads=args.threads,
JVMmemory=args.JVMmemory)
read_utils.rmdup_mvicuna_bam(args.bmtaggerBam,
args.rmdupBam,
JVMmemory=args.JVMmemory)
multi_db_deplete_bam(args.rmdupBam,
args.blastDbs,
deplete_blastn_bam,
args.blastnBam,
threads=args.threads,
JVMmemory=args.JVMmemory)
if args.taxfiltBam and args.lastDb:
filter_lastal_bam(args.blastnBam,
args.lastDb,
args.taxfiltBam,
JVMmemory=args.JVMmemory)
return 0
def main_deplete_human(args):
''' Run the entire depletion pipeline: bmtagger, mvicuna, blastn.
Optionally, use lastal to select a specific taxon of interest.'''
tools.picard.RevertSamTool().execute(
args.inBam, args.revertBam, picardOptions=['SORT_ORDER=queryname', 'SANITIZE=true']
)
multi_db_deplete_bam(
args.revertBam,
args.bmtaggerDbs,
deplete_bmtagger_bam,
args.bmtaggerBam,
threads=args.threads,
JVMmemory=args.JVMmemory
)
read_utils.rmdup_mvicuna_bam(args.bmtaggerBam, args.rmdupBam, JVMmemory=args.JVMmemory)
multi_db_deplete_bam(
args.rmdupBam,
args.blastDbs,
deplete_blastn_bam,
args.blastnBam,
threads=args.threads,
JVMmemory=args.JVMmemory
)
if args.taxfiltBam and args.lastDb:
filter_lastal_bam(args.blastnBam, args.lastDb, args.taxfiltBam, JVMmemory=args.JVMmemory)
return 0
def main_deplete(args):
''' Run the entire depletion pipeline: bwa, bmtagger, mvicuna, blastn.
'''
assert len(args.bmtaggerDbs) + len(args.blastDbs) + len(args.bwaDbs) > 0
# only RevertSam if inBam is already aligned
# Most of the time the input will be unaligned
# so we can save save time if we can skip RevertSam in the unaligned case
#
# via the SAM/BAM spec, if the file is aligned, an SQ line should be present
# in the header. Using pysam, we can check this if header['SQ'])>0
# https://samtools.github.io/hts-specs/SAMv1.pdf
# if the user has requested a revertBam
with read_utils.revert_bam_if_aligned(
args.inBam,
revert_bam=args.revertBam,
clear_tags=args.clear_tags,
tags_to_clear=args.tags_to_clear,
picardOptions=['MAX_DISCARD_FRACTION=0.5'],
JVMmemory=args.JVMmemory,
sanitize=not args.do_not_sanitize) as bamToDeplete:
multi_db_deplete_bam(bamToDeplete,
args.bwaDbs,
deplete_bwa_bam,
args.bwaBam,
threads=args.threads)
def bmtagger_wrapper(inBam, db, outBam, JVMmemory=None):
return deplete_bmtagger_bam(inBam,
db,
outBam,
srprism_memory=args.srprism_memory,
JVMmemory=JVMmemory)
multi_db_deplete_bam(args.bwaBam,
args.bmtaggerDbs,
bmtagger_wrapper,
args.bmtaggerBam,
JVMmemory=args.JVMmemory)
# if the user has not specified saving a revertBam, we used a temp file and can remove it
if not args.revertBam:
os.unlink(revertBamOut)
read_utils.rmdup_mvicuna_bam(args.bmtaggerBam,
args.rmdupBam,
JVMmemory=args.JVMmemory)
multi_db_deplete_bam(args.rmdupBam,
args.blastDbs,
deplete_blastn_bam,
args.blastnBam,
chunkSize=args.chunkSize,
threads=args.threads,
JVMmemory=args.JVMmemory)
return 0
def main_deplete_human(args):
''' Run the entire depletion pipeline: bmtagger, mvicuna, blastn.
Optionally, use lastal to select a specific taxon of interest.'''
# only RevertSam if inBam is already aligned
# Most of the time the input will be unaligned
# so we can save save time if we can skip RevertSam in the unaligned case
#
# via the SAM/BAM spec, if the file is aligned, an SQ line should be present
# in the header. Using pysam, we can check this if header['SQ'])>0
# https://samtools.github.io/hts-specs/SAMv1.pdf
# if the user has requested a revertBam
revertBamOut = args.revertBam if args.revertBam else mkstempfname('.bam')
bamToDeplete = args.inBam
with pysam.AlignmentFile(args.inBam, 'rb', check_sq=False) as bam:
# if it looks like the bam is aligned, revert it
if 'SQ' in bam.header and len(bam.header['SQ'])>0:
tools.picard.RevertSamTool().execute(
args.inBam, revertBamOut, picardOptions=['SORT_ORDER=queryname', 'SANITIZE=true']
)
bamToDeplete = revertBamOut
else:
# if we don't need to produce a revertBam file
# but the user has specified one anyway
# simply touch the output
if args.revertBam:
log.warning("An output was specified for 'revertBam', but the input is unaligned, so RevertSam was not needed. Touching the output.")
util.file.touch(revertBamOut)
# TODO: error out? run RevertSam anyway?
multi_db_deplete_bam(
bamToDeplete,
args.bmtaggerDbs,
deplete_bmtagger_bam,
args.bmtaggerBam,
threads=args.threads,
JVMmemory=args.JVMmemory
)
# if the user has not specified saving a revertBam, we used a temp file and can remove it
if not args.revertBam:
os.unlink(revertBamOut)
read_utils.rmdup_mvicuna_bam(args.bmtaggerBam, args.rmdupBam, JVMmemory=args.JVMmemory)
multi_db_deplete_bam(
args.rmdupBam,
args.blastDbs,
deplete_blastn_bam,
args.blastnBam,
threads=args.threads,
JVMmemory=args.JVMmemory
)
if args.taxfiltBam and args.lastDb:
filter_lastal_bam(args.blastnBam, args.lastDb, args.taxfiltBam, JVMmemory=args.JVMmemory)
return 0
def test_mvicuna_empty_input(self):
samtools = tools.samtools.SamtoolsTool()
empty_bam = os.path.join(util.file.get_test_input_path(), 'empty.bam')
output_bam = util.file.mkstempfname("output.bam")
read_utils.rmdup_mvicuna_bam(
empty_bam,
output_bam
)
self.assertEqual(samtools.count(output_bam), 0)
def main_deplete(args):
''' Run the entire depletion pipeline: bwa, bmtagger, mvicuna, blastn.
'''
assert len(args.bmtaggerDbs) + len(args.blastDbs) + len(args.bwaDbs) > 0
# only RevertSam if inBam is already aligned
# Most of the time the input will be unaligned
# so we can save save time if we can skip RevertSam in the unaligned case
#
# via the SAM/BAM spec, if the file is aligned, an SQ line should be present
# in the header. Using pysam, we can check this if header['SQ'])>0
# https://samtools.github.io/hts-specs/SAMv1.pdf
# if the user has requested a revertBam
with read_utils.revert_bam_if_aligned( args.inBam,
revert_bam = args.revertBam,
clear_tags = args.clear_tags,
tags_to_clear = args.tags_to_clear,
picardOptions = ['MAX_DISCARD_FRACTION=0.5'],
JVMmemory = args.JVMmemory,
sanitize = not args.do_not_sanitize) as bamToDeplete:
multi_db_deplete_bam(
bamToDeplete,
args.bwaDbs,
deplete_bwa_bam,
args.bwaBam,
threads=args.threads
)
def bmtagger_wrapper(inBam, db, outBam, JVMmemory=None):
return deplete_bmtagger_bam(inBam, db, outBam, srprism_memory=args.srprism_memory, JVMmemory=JVMmemory)
multi_db_deplete_bam(
args.bwaBam,
args.bmtaggerDbs,
bmtagger_wrapper,
args.bmtaggerBam,
JVMmemory=args.JVMmemory
)
# if the user has not specified saving a revertBam, we used a temp file and can remove it
if not args.revertBam:
os.unlink(revertBamOut)
read_utils.rmdup_mvicuna_bam(args.bmtaggerBam, args.rmdupBam, JVMmemory=args.JVMmemory)
multi_db_deplete_bam(
args.rmdupBam,
args.blastDbs,
deplete_blastn_bam,
args.blastnBam,
chunkSize=args.chunkSize,
threads=args.threads,
JVMmemory=args.JVMmemory
)
return 0
def test_mvicuna_empty_input(self):
samtools = tools.samtools.SamtoolsTool()
empty_bam = os.path.join(util.file.get_test_input_path(), 'empty.bam')
output_bam = util.file.mkstempfname("output.bam")
read_utils.rmdup_mvicuna_bam(
empty_bam,
output_bam
)
self.assertEqual(samtools.count(output_bam), 0)
def test_mvicuna_canned_input(self):
samtools = tools.samtools.SamtoolsTool()
input_bam = os.path.join(util.file.get_test_input_path(self), 'input.bam')
expected_bam = os.path.join(util.file.get_test_input_path(self), 'expected.bam')
output_bam = util.file.mkstempfname("output.bam")
read_utils.rmdup_mvicuna_bam(
input_bam,
output_bam
)
self.assertEqual(samtools.count(output_bam), samtools.count(expected_bam))
def test_mvicuna_canned_input(self):
samtools = tools.samtools.SamtoolsTool()
input_bam = os.path.join(util.file.get_test_input_path(self), 'input.bam')
expected_bam = os.path.join(util.file.get_test_input_path(self), 'expected.bam')
output_bam = util.file.mkstempfname("output.bam")
read_utils.rmdup_mvicuna_bam(
input_bam,
output_bam
)
self.assertEqual(samtools.count(output_bam), samtools.count(expected_bam))
def main_deplete_human(args):
''' Run the entire depletion pipeline: bmtagger, mvicuna, blastn.
Optionally, use lastal to select a specific taxon of interest.'''
# only RevertSam if inBam is already aligned
# Most of the time the input will be unaligned
# so we can save save time if we can skip RevertSam in the unaligned case
#
# via the SAM/BAM spec, if the file is aligned, an SQ line should be present
# in the header. Using pysam, we can check this if header['SQ'])>0
# https://samtools.github.io/hts-specs/SAMv1.pdf
# if the user has requested a revertBam
revertBamOut = args.revertBam if args.revertBam else mkstempfname('.bam')
bamToDeplete = args.inBam
with pysam.AlignmentFile(args.inBam, 'rb', check_sq=False) as bam:
# if it looks like the bam is aligned, revert it
if 'SQ' in bam.header and len(bam.header['SQ']) > 0:
tools.picard.RevertSamTool().execute(
args.inBam,
revertBamOut,
picardOptions=['SORT_ORDER=queryname', 'SANITIZE=true'])
bamToDeplete = revertBamOut
else:
# if we don't need to produce a revertBam file
# but the user has specified one anyway
# simply touch the output
if args.revertBam:
log.warning(
"An output was specified for 'revertBam', but the input is unaligned, so RevertSam was not needed. Touching the output."
)
util.file.touch(revertBamOut)
# TODO: error out? run RevertSam anyway?
def bmtagger_wrapper(inBam, db, outBam, threads, JVMmemory=None):
return deplete_bmtagger_bam(inBam,
db,
outBam,
threads=threads,
srprism_memory=args.srprism_memory,
JVMmemory=JVMmemory)
multi_db_deplete_bam(bamToDeplete,
args.bmtaggerDbs,
bmtagger_wrapper,
args.bmtaggerBam,
threads=args.threads,
JVMmemory=args.JVMmemory)
# if the user has not specified saving a revertBam, we used a temp file and can remove it
if not args.revertBam:
os.unlink(revertBamOut)
read_utils.rmdup_mvicuna_bam(args.bmtaggerBam,
args.rmdupBam,
JVMmemory=args.JVMmemory)
multi_db_deplete_bam(args.rmdupBam,
args.blastDbs,
deplete_blastn_bam,
args.blastnBam,
chunkSize=args.chunkSize,
threads=args.threads,
JVMmemory=args.JVMmemory)
return 0
def test_mvicuna_empty_input(self):
empty_bam = os.path.join(util.file.get_test_input_path(), 'empty.bam')
read_utils.rmdup_mvicuna_bam(
empty_bam,
util.file.mkstempfname()
)