def main(argv=sys.argv):
parser = E.OptionParser(version="%prog version: $Id: bed2bed.py 2861 2010-02-23 17:36:32Z andreas $",
usage=globals()["__doc__"])
# IMS: new method: extend intervals by set amount
parser.add_option("-m", "--method", dest="methods", type="choice", action="append",
choices=("merge", "filter-genome", "bins",
"block", "sanitize-genome", "shift", "extend"),
help="method to apply [default=%default]")
parser.add_option("--num-bins", dest="num_bins", type="int",
help="number of bins into which to merge (used for method `bins) [default=%default]")
parser.add_option("--bin-edges", dest="bin_edges", type="string",
help="bin_edges for binning method [default=%default]")
parser.add_option("--binning-method", dest="binning_method", type="choice",
choices=(
"equal-bases", "equal-intervals", "equal-range"),
help="method used for binning (used for method `bins` if no bin_edges is given) [default=%default]")
parser.add_option("--merge-distance", dest="merge_distance", type="int",
help="distance in bases over which to merge that are not directly adjacent [default=%default]")
parser.add_option("--merge-min-intervals", dest="merge_min_intervals", type="int",
help="only output merged intervals that are build from at least x intervals [default=%default]")
parser.add_option("--merge-by-name", dest="merge_by_name", action="store_true",
help="only merge intervals with the same name [default=%default]")
parser.add_option("--remove-inconsistent", dest="remove_inconsistent", action="store_true",
help="when merging, do not output intervals where the names of overlapping intervals "
"do not match [default=%default]")
parser.add_option("--offset", dest="offset", type="int",
help="offset for shifting intervals [default=%default]")
parser.add_option("-g", "--genome-file", dest="genome_file", type="string",
help="filename with genome.")
parser.add_option("-b", "--bam-file", dest="bam_file", type="string",
help="bam-formatted filename with genome.")
parser.set_defaults(methods=[],
merge_distance=0,
binning_method="equal-bases",
merge_by_name=False,
genome_file=None,
bam_file=None,
num_bins=5,
merge_min_intervals=1,
bin_edges=None,
offset=10000,
test=None,
extend_distance=1000,
remove_inconsistent=False)
(options, args) = E.Start(parser, add_pipe_options=True)
contigs = None
# Why provide full indexed genome, when a tsv of contig sizes would do?
if options.genome_file:
genome_fasta = IndexedFasta.IndexedFasta(options.genome_file)
contigs = genome_fasta.getContigSizes()
if options.bam_file:
samfile = pysam.Samfile(options.bam_file)
contigs = dict(zip(samfile.references, samfile.lengths))
processor = Bed.iterator(options.stdin)
for method in options.methods:
if method == "filter-genome":
if not contigs:
raise ValueError("please supply contig sizes")
processor = filterGenome(processor, contigs)
elif method == "sanitize-genome":
if not contigs:
raise ValueError("please supply contig sizes")
processor = sanitizeGenome(processor, contigs)
elif method == "merge":
processor = merge(processor,
options.merge_distance,
by_name=options.merge_by_name,
min_intervals=options.merge_min_intervals,
remove_inconsistent=options.remove_inconsistent)
elif method == "bins":
if options.bin_edges:
bin_edges = map(float, options.bin_edges.split(","))
# IMS: check bin edges are valid
if not(len(bin_edges) == options.num_bins + 1):
raise ValueError(
"Number of bin edge must be one more than number of bins")
else:
bin_edges = None
processor, bin_edges = Bed.binIntervals(processor,
num_bins=options.num_bins,
method=options.binning_method,
bin_edges=bin_edges)
E.info("# split bed: bin_edges=%s" % (str(bin_edges)))
elif method == "block":
processor = Bed.blocked_iterator(processor)
elif method == "shift":
# IMS: test that contig sizes are availible
if not contigs:
raise ValueError("please supply genome file")
processor = shiftIntervals(
processor, contigs, offset=options.offset)
# IMS: new method: extend intervals by set amount
elif method == "extend":
if not contigs:
raise ValueError("please supply genome file")
processor = extendInterval(processor, contigs, options.offset)
noutput = 0
for bed in processor:
options.stdout.write(str(bed) + "\n")
noutput += 1
E.info("noutput=%i" % (noutput))
E.Stop()
def main(argv=sys.argv):
parser = E.OptionParser(
version=
"%prog version: $Id: bed2bed.py 2861 2010-02-23 17:36:32Z andreas $",
usage=globals()["__doc__"])
# IMS: new method: extend intervals by set amount
parser.add_option("-m",
"--method",
dest="methods",
type="choice",
action="append",
choices=("merge", "filter-genome", "bins", "block",
"sanitize-genome", "shift", "extend",
"filter-names"),
help="method to apply [default=%default]")
parser.add_option("--num-bins",
dest="num_bins",
type="int",
help="number of bins into which to merge (used for "
"method `bins) [default=%default]")
parser.add_option("--bin-edges",
dest="bin_edges",
type="string",
help="bin_edges for binning method [default=%default]")
parser.add_option(
"--binning-method",
dest="binning_method",
type="choice",
choices=("equal-bases", "equal-intervals", "equal-range"),
help="method used for binning (used for method `bins` if no "
"bin_edges is given) [default=%default]")
parser.add_option(
"--merge-distance",
dest="merge_distance",
type="int",
help="distance in bases over which to merge that are not "
"directly adjacent [default=%default]")
parser.add_option(
"--merge-min-intervals",
dest="merge_min_intervals",
type="int",
help="only output merged intervals that are build from at least "
"x intervals [default=%default]")
parser.add_option(
"--merge-by-name",
dest="merge_by_name",
action="store_true",
help="only merge intervals with the same name [default=%default]")
parser.add_option(
"--merge-and-resolve-blocks",
dest="resolve_blocks",
action="store_true",
help="When merging bed12 entrys, should blocks be resolved?")
parser.add_option("--merge-stranded",
dest="stranded",
action="store_true",
help="Only merge intervals on the same strand")
parser.add_option(
"--remove-inconsistent-names",
dest="remove_inconsistent_names",
action="store_true",
help="when merging, do not output intervals where the names of "
"overlapping intervals do not match [default=%default]")
parser.add_option("--offset",
dest="offset",
type="int",
help="offset for shifting intervals [default=%default]")
parser.add_option("-g",
"--genome-file",
dest="genome_file",
type="string",
help="filename with genome.")
parser.add_option("-b",
"--bam-file",
dest="bam_file",
type="string",
help="bam-formatted filename with genome.")
parser.add_option("--filter-names-file",
dest="names",
type="string",
help="list of names to keep. One per line")
parser.set_defaults(methods=[],
merge_distance=0,
binning_method="equal-bases",
merge_by_name=False,
genome_file=None,
bam_file=None,
num_bins=5,
merge_min_intervals=1,
bin_edges=None,
offset=10000,
test=None,
extend_distance=1000,
remove_inconsistent_names=False,
resolve_blocks=False)
(options, args) = E.Start(parser, add_pipe_options=True)
contigs = None
# Why provide full indexed genome, when a tsv of contig sizes would do?
if options.genome_file:
genome_fasta = IndexedFasta.IndexedFasta(options.genome_file)
contigs = genome_fasta.getContigSizes()
if options.bam_file:
samfile = pysam.AlignmentFile(options.bam_file)
contigs = dict(list(zip(samfile.references, samfile.lengths)))
processor = Bed.iterator(options.stdin)
for method in options.methods:
if method == "filter-genome":
if not contigs:
raise ValueError("please supply contig sizes")
processor = filterGenome(processor, contigs)
elif method == "sanitize-genome":
if not contigs:
raise ValueError("please supply contig sizes")
processor = sanitizeGenome(processor, contigs)
elif method == "merge":
processor = merge(
processor,
options.merge_distance,
by_name=options.merge_by_name,
min_intervals=options.merge_min_intervals,
remove_inconsistent=options.remove_inconsistent_names,
resolve_blocks=options.resolve_blocks,
stranded=options.stranded)
elif method == "bins":
if options.bin_edges:
bin_edges = list(map(float, options.bin_edges.split(",")))
# IMS: check bin edges are valid
if not (len(bin_edges) == options.num_bins + 1):
raise ValueError(
"Number of bin edge must be one more than "
"number of bins")
else:
bin_edges = None
processor, bin_edges = Bed.binIntervals(
processor,
num_bins=options.num_bins,
method=options.binning_method,
bin_edges=bin_edges)
E.info("# split bed: bin_edges=%s" % (str(bin_edges)))
elif method == "block":
processor = Bed.blocked_iterator(processor)
elif method == "shift":
# IMS: test that contig sizes are availible
if not contigs:
raise ValueError("please supply genome file")
processor = shiftIntervals(processor,
contigs,
offset=options.offset)
# IMS: new method: extend intervals by set amount
elif method == "extend":
if not contigs:
raise ValueError("please supply genome file")
processor = extendInterval(processor, contigs, options.offset)
elif method == "filter-names":
if not options.names:
raise ValueError("please supply list of names to filter")
names = [name.strip() for name in open(options.names)]
processor = filterNames(processor, names)
noutput = 0
for bed in processor:
options.stdout.write(str(bed) + "\n")
noutput += 1
E.info("noutput=%i" % (noutput))
E.Stop()
