def get_parser():
"""
returns the parser object for the oxli subcommand handler
"""
parser = argparse.ArgumentParser(
description="Single entry point script for khmer", formatter_class=argparse.ArgumentDefaultsHelpFormatter
)
subparsers = parser.add_subparsers()
# build-graph (formerly load-graph) parsers here
parser_build_graph = subparsers.add_parser(
"build-graph",
help="Load sequences into the compressible graph" "format plus optional tagset",
description="Load sequences into the " "compressible graph format plus optional tagset",
)
khmer_args.build_hashbits_args(
"Load sequences into the compressible" "graph format plus optional tagset.", None, parser=parser_build_graph
)
build_graph.build_parser(parser_build_graph)
parser_build_graph.set_defaults(func=build_graph.main)
return parser
def get_parser():
parser = build_hashbits_args(descr="Load sequences into the compressible "
"graph format plus optional tagset.")
add_threading_args(parser)
parser.add_argument('--no-build-tagset',
'-n',
default=False,
action='store_true',
dest='no_build_tagset',
help='Do NOT construct tagset while loading sequences')
parser.add_argument('output_filename',
metavar='output_presence_table_filename',
help='output'
' k-mer presence table filename.')
parser.add_argument('input_filenames',
metavar='input_sequence_filename',
nargs='+',
help='input FAST[AQ] sequence filename')
parser.add_argument('--report-total-kmers',
'-t',
action='store_true',
help="Prints the total number of k-mers to stderr")
parser.add_argument('--write-fp-rate',
'-w',
action='store_true',
help="Write false positive rate into .info file")
return parser
def get_parser():
epilog = """
An additional report will be written to ${output_report_filename}.curve
containing the increase of overlap k-mers as the number of sequences in the
second database increases.
"""
parser = build_hashbits_args(
descr='Count the overlap k-mers which are the k-mers appearing in two '
'sequence datasets.',
epilog=textwrap.dedent(epilog))
parser.add_argument('ptfile',
metavar='input_presence_table_filename',
help="input k-mer presence table filename")
parser.add_argument('fafile',
metavar='input_sequence_filename',
help="input sequence filename")
parser.add_argument('report_filename',
metavar='output_report_filename',
help='output report filename')
parser.add_argument('--csv',
default=False,
action='store_true',
help='Use the CSV format for the curve output. '
'Includes column headers.')
parser.add_argument('-f',
'--force',
default=False,
action='store_true',
help='Overwrite output file if it exists')
return parser
def get_parser():
epilog = """
Load in a set of sequences, partition them, merge the partitions, and
annotate the original sequences files with the partition information.
This script combines the functionality of :program:`load-graph.py`,
:program:`partition-graph.py`, :program:`merge-partitions.py`, and
:program:`annotate-partitions.py` into one script. This is convenient
but should probably not be used for large data sets, because
:program:`do-partition.py` doesn't provide save/resume functionality.
"""
parser = build_hashbits_args(
descr='Load, partition, and annotate FAST[AQ] sequences',
epilog=textwrap.dedent(epilog))
add_threading_args(parser)
parser.add_argument('--subset-size', '-s', default=DEFAULT_SUBSET_SIZE,
dest='subset_size', type=float,
help='Set subset size (usually 1e5-1e6 is good)')
parser.add_argument('--no-big-traverse', dest='no_big_traverse',
action='store_true', default=False,
help='Truncate graph joins at big traversals')
parser.add_argument('--keep-subsets', dest='remove_subsets',
default=True, action='store_false',
help='Keep individual subsets (default: False)')
parser.add_argument('graphbase', help="base name for output files")
parser.add_argument('input_filenames', metavar='input_sequence_filename',
nargs='+', help='input FAST[AQ] sequence filenames')
parser.add_argument('-f', '--force', default=False, action='store_true',
help='Overwrite output file if it exists')
return parser
def get_parser():
epilog = """
Load in a set of sequences, partition them, merge the partitions, and
annotate the original sequences files with the partition information.
This script combines the functionality of :program:`load-graph.py`,
:program:`partition-graph.py`, :program:`merge-partitions.py`, and
:program:`annotate-partitions.py` into one script. This is convenient
but should probably not be used for large data sets, because
:program:`do-partition.py` doesn't provide save/resume functionality.
"""
parser = build_hashbits_args(
descr='Load, partition, and annotate FAST[AQ] sequences',
epilog=textwrap.dedent(epilog))
parser.add_argument('--subset-size', '-s', default=DEFAULT_SUBSET_SIZE,
dest='subset_size', type=float,
help='Set subset size (usually 1e5-1e6 is good)')
parser.add_argument('--no-big-traverse', dest='no_big_traverse',
action='store_true', default=False,
help='Truncate graph joins at big traversals')
parser.add_argument('--threads', '-T', dest='n_threads',
default=DEFAULT_N_THREADS,
help='Number of simultaneous threads to execute')
parser.add_argument('--keep-subsets', dest='remove_subsets',
default=True, action='store_false',
help='Keep individual subsets (default: False)')
parser.add_argument('graphbase', help="base name for output files")
parser.add_argument('input_filenames', metavar='input_sequence_filename',
nargs='+', help='input FAST[AQ] sequence filenames')
return parser
def get_parser():
parser = build_hashbits_args(
"Takes a partitioned reference file \
and a list of reads, and sorts reads \
by which partition they connect to"
)
parser.epilog = EPILOG
parser.add_argument(
"-r",
"--traversal_range",
type=int,
dest="traversal_range",
default=DEFAULT_RANGE,
help="depth of breadth-first search to perform\
from each read",
)
parser.add_argument("--max_queue_size", type=int, default=1000)
parser.add_argument("--prefix", dest="output_prefix", default=DEFAULT_OUT_PREF, help="Prefix for sorted read files")
parser.add_argument(
"--outdir",
dest="outdir",
default="",
help="output directory; default is location of \
fastp file",
)
parser.add_argument("--query", dest="query", nargs="+", help="Reads to be swept and sorted")
parser.add_argument("--db", dest="db", nargs="+", help="Database reads for sweep", required=True)
return parser
def get_parser():
parser = build_hashbits_args('Takes a partitioned reference file \
and a list of reads, and sorts reads \
by which partition they connect to')
parser.epilog = EPILOG
parser.add_argument('-r',
'--traversal_range',
type=int,
dest='traversal_range',
default=DEFAULT_RANGE,
help='depth of breadth-first search to perform\
from each read')
parser.add_argument('--max_queue_size', type=int, default=1000)
parser.add_argument('--prefix',
dest='output_prefix',
default=DEFAULT_OUT_PREF,
help='Prefix for sorted read files')
parser.add_argument('--outdir',
dest='outdir',
default='',
help='output directory; default is location of \
fastp file')
parser.add_argument('--query',
dest='query',
nargs='+',
help='Reads to be swept and sorted')
parser.add_argument('--db',
dest='db',
nargs='+',
help='Database reads for sweep',
required=True)
return parser
def get_parser():
parser = build_hashbits_args(descr="Load sequences into the compressible "
"graph format plus optional tagset.")
add_threading_args(parser)
parser.add_argument('input_filenames', metavar='input_sequence_filename',
nargs='+', help='input FAST[AQ] sequence filename')
return parser
def get_parser():
parser = build_hashbits_args(descr="Load sequences into the compressible "
"graph format plus optional tagset.")
add_threading_args(parser)
parser.add_argument('input_filenames',
metavar='input_sequence_filename',
nargs='+',
help='input FAST[AQ] sequence filename')
return parser
def get_parser():
parser = build_hashbits_args(descr="Load sequences into the compressible "
"graph format plus optional tagset.")
add_threading_args(parser)
parser.add_argument('--no-build-tagset', '-n', default=False,
action='store_true', dest='no_build_tagset',
help='Do NOT construct tagset while loading sequences')
parser.add_argument('output_filename',
metavar='output_presence_table_filename', help='output'
' k-mer presence table filename.')
parser.add_argument('input_filenames', metavar='input_sequence_filename',
nargs='+', help='input FAST[AQ] sequence filename')
return parser
def get_parser():
epilog = """
An additional report will be written to ${output_report_filename}.curve
containing the increase of overlap k-mers as the number of sequences in the
second database increases.
"""
parser = build_hashbits_args(
descr='Count the overlap k-mers which are the k-mers appearing in two '
'sequence datasets.', epilog=textwrap.dedent(epilog))
parser.add_argument('ptfile', metavar='input_presence_table_filename',
help="input k-mer presence table filename")
parser.add_argument('fafile', metavar='input_sequence_filename',
help="input sequence filename")
parser.add_argument('report_filename', metavar='output_report_filename',
help='output report filename')
return parser
def get_parser():
parser = build_hashbits_args(descr="Load sequences into the compressible "
"graph format plus optional tagset.")
add_threading_args(parser)
parser.add_argument('--no-build-tagset', '-n', default=False,
action='store_true', dest='no_build_tagset',
help='Do NOT construct tagset while loading sequences')
parser.add_argument('output_filename',
metavar='output_presence_table_filename', help='output'
' k-mer presence table filename.')
parser.add_argument('input_filenames', metavar='input_sequence_filename',
nargs='+', help='input FAST[AQ] sequence filename')
parser.add_argument('--report-total-kmers', '-t', action='store_true',
help="Prints the total number of k-mers to stderr")
parser.add_argument('--write-fp-rate', '-w', action='store_true',
help="Write false positive rate into .info file")
return parser
def get_parser():
parser = build_hashbits_args(descr="Load sequences into the compressible "
"graph format plus optional tagset.")
add_threading_args(parser)
parser.add_argument('--no-build-tagset',
'-n',
default=False,
action='store_true',
dest='no_build_tagset',
help='Do NOT construct tagset while loading sequences')
parser.add_argument('output_filename',
metavar='output_presence_table_filename',
help='output'
' k-mer presence table filename.')
parser.add_argument('input_filenames',
metavar='input_sequence_filename',
nargs='+',
help='input FAST[AQ] sequence filename')
return parser
def get_parser():
epilog = """
Load in a set of sequences, partition them, merge the partitions, and
annotate the original sequences files with the partition information.
This script combines the functionality of :program:`load-graph.py`,
:program:`partition-graph.py`, :program:`merge-partitions.py`, and
:program:`annotate-partitions.py` into one script. This is convenient
but should probably not be used for large data sets, because
:program:`do-partition.py` doesn't provide save/resume functionality.
"""
parser = build_hashbits_args(
descr="Load, partition, and annotate FAST[AQ] sequences", epilog=textwrap.dedent(epilog)
)
add_threading_args(parser)
parser.add_argument(
"--subset-size",
"-s",
default=DEFAULT_SUBSET_SIZE,
dest="subset_size",
type=float,
help="Set subset size (usually 1e5-1e6 is good)",
)
parser.add_argument(
"--no-big-traverse",
dest="no_big_traverse",
action="store_true",
default=False,
help="Truncate graph joins at big traversals",
)
parser.add_argument(
"--keep-subsets",
dest="remove_subsets",
default=True,
action="store_false",
help="Keep individual subsets (default: False)",
)
parser.add_argument("graphbase", help="base name for output files")
parser.add_argument(
"input_filenames", metavar="input_sequence_filename", nargs="+", help="input FAST[AQ] sequence filenames"
)
parser.add_argument("-f", "--force", default=False, action="store_true", help="Overwrite output file if it exists")
return parser
def get_parser():
parser = build_hashbits_args('Takes a partitioned reference file \
and a list of reads, and sorts reads \
by which partition they connect to')
parser.epilog = EPILOG
parser.add_argument(
'-r', '--traversal_range', type=int, dest='traversal_range',
default=DEFAULT_RANGE, help='depth of breadth-first search to perform\
from each read')
parser.add_argument('-b', '--buffer_size', dest='max_reads', type=int,
default=DEFAULT_MAX_READS,
help='Max total reads to buffer before flushing')
parser.add_argument('-l', '--buffer_length', dest='buffer_size', type=int,
default=DEFAULT_BUFFER_SIZE,
help='Max length of an individual label buffer \
before flushing')
parser.add_argument('--prefix', dest='output_prefix',
default=DEFAULT_OUT_PREF,
help='Prefix for sorted read files')
parser.add_argument('--outdir', dest='outdir',
help='output directory; default is location of \
fastp file')
parser.add_argument('-m', '--max_buffers', dest='max_buffers', type=int,
default=DEFAULT_NUM_BUFFERS,
help='Max individual label buffers before flushing')
labeling = parser.add_mutually_exclusive_group(required=True)
labeling.add_argument('--label-by-pid', dest='label_by_pid',
action='store_true', help='separate reads by\
reference partition id')
labeling.add_argument('--label-by-seq', dest='label_by_seq',
action='store_true', help='separate reads by\
reference sequence')
labeling.add_argument('--label-by-group', dest='group_size', type=int,
help='separate reads by arbitrary sized groups\
of reference sequences')
parser.add_argument(dest='input_fastp', help='Reference fasta or fastp')
parser.add_argument('input_files', nargs='+',
help='Reads to be swept and sorted')
parser.add_argument('-f', '--force', default=False, action='store_true',
help='Overwrite output file if it exists')
return parser
def get_parser():
parser = build_hashbits_args('Takes a partitioned reference file \
and a list of reads, and sorts reads \
by which partition they connect to')
parser.epilog = EPILOG
parser.add_argument(
'-r', '--traversal_range', type=int, dest='traversal_range',
default=DEFAULT_RANGE, help='depth of breadth-first search to perform\
from each read')
parser.add_argument('-b', '--buffer_size', dest='max_reads', type=int,
default=DEFAULT_MAX_READS,
help='Max total reads to buffer before flushing')
parser.add_argument('-l', '--buffer_length', dest='buffer_size', type=int,
default=DEFAULT_BUFFER_SIZE,
help='Max length of an individual label buffer \
before flushing')
parser.add_argument('--prefix', dest='output_prefix',
default=DEFAULT_OUT_PREF,
help='Prefix for sorted read files')
parser.add_argument('--outdir', dest='outdir',
help='output directory; default is location of \
fastp file')
parser.add_argument('-m', '--max_buffers', dest='max_buffers', type=int,
default=DEFAULT_NUM_BUFFERS,
help='Max individual label buffers before flushing')
labeling = parser.add_mutually_exclusive_group(required=True)
labeling.add_argument('--label-by-pid', dest='label_by_pid',
action='store_true', help='separate reads by\
referece partition id')
labeling.add_argument('--label-by-seq', dest='label_by_seq',
action='store_true', help='separate reads by\
reference sequence')
labeling.add_argument('--label-by-group', dest='group_size', type=int,
help='separate reads by arbitrary sized groups\
of reference sequences')
parser.add_argument(dest='input_fastp', help='Reference fasta or fastp')
parser.add_argument('input_files', nargs='+',
help='Reads to be swept and sorted')
parser.add_argument('-f', '--force', default=False, action='store_true',
help='Overwrite output file if it exists')
return parser
def get_parser():
epilog = """
An additional report will be written to ${output_report_filename}.curve
containing the increase of overlap k-mers as the number of sequences in the
second database increases.
"""
parser = build_hashbits_args(
descr='Count the overlap k-mers which are the k-mers appearing in two '
'sequence datasets.', epilog=textwrap.dedent(epilog))
parser.add_argument('ptfile', metavar='input_presence_table_filename',
help="input k-mer presence table filename")
parser.add_argument('fafile', metavar='input_sequence_filename',
help="input sequence filename")
parser.add_argument('report_filename', metavar='output_report_filename',
help='output report filename')
parser.add_argument('--csv', default=False, action='store_true',
help='Use the CSV format for the curve output. '
'Includes column headers.')
parser.add_argument('-f', '--force', default=False, action='store_true',
help='Overwrite output file if it exists')
return parser
def get_parser():
epilog = """
An additional report will be written to ${output_report_filename}.curve
containing the increase of overlap k-mers as the number of sequences in the
second database increases.
"""
parser = build_hashbits_args(
descr='Count the overlap k-mers which are the k-mers appearing in two '
'sequence datasets.',
epilog=textwrap.dedent(epilog))
parser.add_argument('ptfile',
metavar='input_presence_table_filename',
help="input k-mer presence table filename")
parser.add_argument('fafile',
metavar='input_sequence_filename',
help="input sequence filename")
parser.add_argument('report_filename',
metavar='output_report_filename',
help='output report filename')
return parser
def get_parser():
parser = build_hashbits_args(descr="Load sequences into the compressible " "graph format plus optional tagset.")
add_threading_args(parser)
parser.add_argument(
"--no-build-tagset",
"-n",
default=False,
action="store_true",
dest="no_build_tagset",
help="Do NOT construct tagset while loading sequences",
)
parser.add_argument(
"output_filename", metavar="output_presence_table_filename", help="output" " k-mer presence table filename."
)
parser.add_argument(
"input_filenames", metavar="input_sequence_filename", nargs="+", help="input FAST[AQ] sequence filename"
)
parser.add_argument(
"--report-total-kmers", "-t", action="store_true", help="Prints the total number of k-mers to stderr"
)
parser.add_argument("--write-fp-rate", "-w", action="store_true", help="Write false positive rate into .info file")
return parser
def get_parser():
parser = build_hashbits_args('Takes a partitioned reference file \
and a list of reads, and sorts reads \
by which partition they connect to')
parser.epilog = EPILOG
parser.add_argument(
'-r', '--traversal_range', type=int, dest='traversal_range',
default=DEFAULT_RANGE, help='depth of breadth-first search to perform\
from each read')
parser.add_argument('--max_queue_size', type=int, default=1000)
parser.add_argument('--prefix', dest='output_prefix',
default=DEFAULT_OUT_PREF,
help='Prefix for sorted read files')
parser.add_argument('--outdir', dest='outdir', default='',
help='output directory; default is location of \
fastp file')
parser.add_argument('--query', dest='query', nargs='+',
help='Reads to be swept and sorted')
parser.add_argument('--db', dest='db', nargs='+',
help='Database reads for sweep', required=True)
return parser
def get_parser():
parser = build_hashbits_args(descr="Load sequences into the compressible "
"graph format plus optional tagset.")
parser = build_graph.build_parser(parser)
return parser
def get_parser():
parser = build_hashbits_args(
"Takes a partitioned reference file \
and a list of reads, and sorts reads \
by which partition they connect to"
)
parser.epilog = EPILOG
parser.add_argument(
"-r",
"--traversal_range",
type=int,
dest="traversal_range",
default=DEFAULT_RANGE,
help="depth of breadth-first search to perform\
from each read",
)
parser.add_argument(
"-b",
"--buffer_size",
dest="max_reads",
type=int,
default=DEFAULT_MAX_READS,
help="Max total reads to buffer before flushing",
)
parser.add_argument(
"-l",
"--buffer_length",
dest="buffer_size",
type=int,
default=DEFAULT_BUFFER_SIZE,
help="Max length of an individual label buffer \
before flushing",
)
parser.add_argument("--prefix", dest="output_prefix", default=DEFAULT_OUT_PREF, help="Prefix for sorted read files")
parser.add_argument(
"--outdir",
dest="outdir",
help="output directory; default is location of \
fastp file",
)
parser.add_argument(
"-m",
"--max_buffers",
dest="max_buffers",
type=int,
default=DEFAULT_NUM_BUFFERS,
help="Max individual label buffers before flushing",
)
labeling = parser.add_mutually_exclusive_group(required=True)
labeling.add_argument(
"--label-by-pid",
dest="label_by_pid",
action="store_true",
help="separate reads by\
referece partition id",
)
labeling.add_argument(
"--label-by-seq",
dest="label_by_seq",
action="store_true",
help="separate reads by\
reference sequence",
)
labeling.add_argument(
"--label-by-group",
dest="group_size",
type=int,
help="separate reads by arbitrary sized groups\
of reference sequences",
)
parser.add_argument(dest="input_fastp", help="Reference fasta or fastp")
parser.add_argument("input_files", nargs="+", help="Reads to be swept and sorted")
return parser
def get_parser():
parser = build_hashbits_args(descr="Load sequences into the compressible "
"graph format plus optional tagset.")
parser = build_graph.build_parser(parser)
return parser
def main():
parser = build_hashbits_args()
parser.add_argument('input_filenames', nargs='+')
parser.add_argument('read_filename')
args = parser.parse_args()
if not args.quiet:
if args.min_tablesize == DEFAULT_MIN_TABLESIZE:
print >>sys.stderr, "** WARNING: tablesize is default! " \
"You absodefly want to increase this!\n** " \
"Please read the docs!"
print >>sys.stderr, '\nPARAMETERS:'
print >>sys.stderr, ' - kmer size = %d \t\t(-k)' % args.ksize
print >>sys.stderr, ' - n hashes = %d \t\t(-N)' % args.n_tables
print >>sys.stderr, ' - min hashsize = %-5.2g \t(-x)' % \
args.min_tablesize
print >>sys.stderr, ''
print >>sys.stderr, 'Estimated memory usage is %.2g bytes ' \
'(n_tables x min_tablesize / 8)' % (
args.n_tables * args.min_tablesize * len(args.input_filenames) / 8.)
print >>sys.stderr, '-' * 8
K = args.ksize
HT_SIZE = args.min_tablesize
N_HT = args.n_tables
inputlist = args.input_filenames
readsfile = args.read_filename
query_list = []
for n, inp_name in enumerate(inputlist):
# create a hashbits data structure
ht = khmer.new_hashbits(K, HT_SIZE, N_HT)
outfile = os.path.basename(inp_name) + '.sweep3'
outfp = open(outfile, 'w')
query_list.append((ht, outfp))
for n, inp_name in enumerate(inputlist):
ht = query_list[n][0]
# load contigs, connect into N partitions
print 'loading input reads from', inp_name
ht.consume_fasta(inp_name)
print 'starting sweep.'
n = 0
m = 0
for n, record in enumerate(screed.open(readsfile)):
if len(record.sequence) < K:
continue
if n % 10000 == 0:
print '...', n, m
for ht, outfp in query_list:
count = ht.get_median_count(record.sequence)[0]
if count:
outfp.write(output_single(record))
