def main():
p = optparse.OptionParser(__doc__)
p.add_option("-a", dest="a", help="file to annotate. first 3 columns are "
"chrom start stop")
p.add_option("-b", dest="b", help="superbed to annotate with")
p.add_option("--header", dest="header", help="a file has a header",
action="store_true", default=False)
p.add_option("-N","--no-near", dest="no_near",
help="dont find the nearest gene, just the up/downstream",
action="store_true", default=False)
p.add_option("--upstream", dest="upstream", type=int, default=None,
help="distance upstream of [a] to look for [b]")
p.add_option("--downstream", dest="downstream", type=int, default=None,
help="distance downstream of [a] to look for [b]")
p.add_option("--transcripts", dest="transcripts", action="store_true",
default=False, help="use transcript names in output as well as"
" gene name. default is just gene name")
opts, args = p.parse_args()
if (opts.a is None or opts.b is None):
sys.exit(not p.print_help())
b = opts.b
if not opts.transcripts:
b = remove_transcripts(b)
if not (opts.upstream or opts.downstream):
superanno(opts.a, b, opts.header, opts.no_near, sys.stdout)
else:
out = open(BedTool._tmp(), "w")
superanno(opts.a, b, opts.header, opts.no_near, out)
out.close()
new_header = []
out_fh = open(out.name)
new_header = [out_fh.readline().rstrip("\r\n")] if opts.header else []
for xdir in ("upstream", "downstream"):
dist = getattr(opts, xdir)
if dist is None: continue
new_out = open(BedTool._tmp(), "w")
xstream(out_fh, b, dist, xdir, new_out)
new_header.append("%s_%i" % (xdir, dist))
new_out.close()
out_fh = open(new_out.name)
if opts.header:
print "\t".join(new_header)
for line in open(out_fh.name):
sys.stdout.write(line)
def nearest(a, b):
a_not_overlapping = a.intersect(b, v=True)
if len(a_not_overlapping) != 0:
ab = a_not_overlapping.closest(b, t="all", stream=True)
else:
ab = []
by_name = collections.defaultdict(list)
for row in ab:
key = row[3]
row[3] = "."
by_name[key].append(row)
fh = open(BedTool._tmp(), "w")
seen = set()
for name, rows in by_name.iteritems():
# TODO: just like above.
full_names = [r[9] for r in rows]
dists = [get_dist(r) for r in rows]
if len(set(dists)) == 1:
dists = set(dists)
full_names = set(full_names)
dists = ";".join(map(str, dists))
names = ";".join(full_names)
line = "\t".join(name.split("Z_Z") + [names, dists])
if line in seen: continue
seen.add(line)
fh.write(line + "\n")
fh.close()
return fh.name
def nearest(a, b, bstrand=None):
a_not_overlapping = a.intersect(b, v=True)
if len(a_not_overlapping) != 0:
ab = a_not_overlapping.closest(b, t="all", stream=True)
else:
ab = []
by_name = collections.defaultdict(list)
for row in ab:
key = row[3]
row[3] = "."
by_name[key].append(row)
fh = open(BedTool._tmp(), "w")
seen = set()
for name, rows in by_name.iteritems():
# TODO: just like above.
full_names = [r[9] for r in rows]
dists = [get_dist(r, bstrand) for r in rows]
if len(set(dists)) == 1:
dists = set(dists)
full_names = set(full_names)
dists = ";".join(map(str, dists))
names = ";".join(full_names)
line = "\t".join(name.split("Z_Z") + [names, dists])
if line in seen: continue
seen.add(line)
fh.write(line + "\n")
fh.close()
return fh.name
def add_xstream(a, b, dist, updown, report_distance=False):
# run a window up or downstream.
dir = dict(up="l", down="r")[updown]
kwargs = {'sw':True, dir: dist}
# have to set the other to 0
if "l" in kwargs: kwargs["r"] = 0
else: kwargs["l"] = 0
c = a.window(b, **kwargs)
afields = a.field_count()
get_name = gen_get_name(b, afields)
seen = collections.defaultdict(set)
# condense to unique names.
for row in c:
key = "\t".join(row[:afields])
seen[key].update([get_name(row)])
d = open(BedTool._tmp(), "w")
for row in seen:
d.write(row + "\t" + ",".join(sorted(seen[row])) + "\n")
# write the entries that did not appear in the window'ed Bed
for row in a:
key = "\t".join(row[:afields])
if key in seen: continue
d.write(str(row) + "\t.\n")
d.close()
dbed = BedTool(d.name)
assert len(dbed) == len(a)
return dbed
def simplify_bed(fbed, has_header):
"""
create a bed with no header and 6 columns.
retain strand info.
"""
line_gen = reader(fbed, header=False)
header = line_gen.next() if has_header else None
fh = open(BedTool._tmp(), "w")
for toks in line_gen:
new_toks = toks[:3] + ["Z_Z".join(toks), ".",
toks[5] if len(toks) > 5 else "."]
fh.write("\t".join(new_toks) + "\n")
fh.close()
return BedTool(fh.name), header
def simplify_bed(fbed, has_header):
"""
create a bed with no header and 6 columns.
retain strand info.
"""
line_gen = reader(fbed, header=False)
header = line_gen.next() if has_header else None
fh = open(BedTool._tmp(), "w")
for toks in line_gen:
new_toks = toks[:3] + [
"Z_Z".join(toks), ".", toks[5] if len(toks) > 5 else "."
]
fh.write("\t".join(new_toks) + "\n")
fh.close()
return BedTool(fh.name), header
def overlapping(a, b):
by_name = collections.defaultdict(list)
for row in a.intersect(b, wo=True, stream=True).cut(range(6) + [9, 10],
stream=True):
key = row[3] # the ZZ joined string.
# 6, 7 are name, type.
by_name[key].append((row[6], row[7]))
fh = open(BedTool._tmp(), "w")
for name, rows in by_name.iteritems():
types = sorted(set([r[1] for r in rows]))
full_names = sorted(set([r[0] for r in rows]))
#regain the original line.
line = name.split("Z_Z") + [";".join(full_names), ";".join(types)]
fh.write("\t".join(line) + "\n")
fh.close()
return fh.name
def overlapping(a, b):
by_name = collections.defaultdict(list)
for row in a.intersect(b, wo=True, stream=True).cut(range(6) + [9, 10],
stream=True):
key = row[3] # the ZZ joined string.
# 6, 7 are name, type.
by_name[key].append((row[6], row[7]))
fh = open(BedTool._tmp(), "w")
for name, rows in by_name.iteritems():
types = sorted(set([r[1] for r in rows]))
full_names = sorted(set([r[0] for r in rows]))
#regain the original line.
line = name.split("Z_Z") + [";".join(full_names), ";".join(types)]
fh.write("\t".join(line) + "\n")
fh.close()
return fh.name
def remove_transcripts(b):
if op.exists(b + ".notranscripts"):
if os.stat(b + ".notranscripts").st_mtime > os.stat(b).st_mtime:
return b + ".notranscripts"
try:
bnew = open(b + ".notranscripts", 'w')
except:
bnew = open(BedTool._tmp(), "w")
try:
for row in reader(b, header=False):
if "," in row[3]:
row[3] = row[3].split(",")[1]
bnew.write("\t".join(row) + "\n")
bnew.close()
except:
os.unlink(bnew.name)
raise
return bnew.name
def remove_transcripts(b):
if op.exists(b + ".notranscripts"):
if os.stat(b + ".notranscripts").st_mtime > os.stat(b).st_mtime:
return b + ".notranscripts"
try:
bnew = open(b + ".notranscripts", 'w')
except:
bnew = open(BedTool._tmp(), "w")
try:
for row in reader(b, header=False):
if "," in row[3]:
row[3] = row[3].split(",")[1]
bnew.write("\t".join(row) + "\n")
bnew.close()
except:
os.unlink(bnew.name)
raise
return bnew.name
def add_xstream(a, b, dist, updown, report_distance=False):
# run a window up or downstream.
dir = dict(up="l", down="r")[updown]
kwargs = {"sw": True, dir: dist}
# have to set the other to 0
if "l" in kwargs:
kwargs["r"] = 0
else:
kwargs["l"] = 0
c = a.window(b, **kwargs)
afields = a.field_count()
get_name = gen_get_name(b, afields)
seen = collections.defaultdict(set)
# condense to unique names.
for row in c:
key = "\t".join(row[:afields])
seen[key].update([get_name(row)])
d = open(BedTool._tmp(), "w")
for row in seen:
d.write(row + "\t" + ",".join(sorted(seen[row])) + "\n")
# write the entries that did not appear in the window'ed Bed
for row in a:
key = "\t".join(row[:afields])
if key in seen:
continue
d.write(str(row) + "\t.\n")
d.close()
dbed = BedTool(d.name)
assert len(dbed) == len(a)
return dbed
def add_closest(aname, bname):
a, b = BedTool(aname), BedTool(bname)
afields = a.field_count()
c = a.closest(b, d=True)
get_name = gen_get_name(b, afields)
dbed = open(BedTool._tmp(), "w")
# keep the name and distance
seen_by_line = collections.defaultdict(list)
for feat in c:
key = "\t".join(feat[:afields])
seen_by_line[key].append([feat[-1], get_name(feat)])
for key, dist_names in seen_by_line.items():
if len(dist_names) > 0:
assert len(set([d[0] for d in dist_names])) == 1
names = ",".join(sorted(set(d[1] for d in dist_names)))
new_line = "\t".join([key] + [names] + [dist_names[0][0]])
dbed.write(new_line + "\n")
dbed.close()
d = BedTool(dbed.name)
assert len(d) == len(a)
return d
def add_closest(aname, bname):
a, b = BedTool(aname), BedTool(bname)
afields = a.field_count()
c = a.closest(b, d=True)
get_name = gen_get_name(b, afields)
dbed = open(BedTool._tmp(), "w")
# keep the name and distance
seen_by_line = collections.defaultdict(list)
for feat in c:
key = "\t".join(feat[:afields])
seen_by_line[key].append([feat[-1], get_name(feat)])
for key, dist_names in seen_by_line.iteritems():
if len(dist_names) > 0:
assert len(set([d[0] for d in dist_names])) == 1
names = ",".join(sorted(set(d[1] for d in dist_names)))
new_line = "\t".join([key] + [names] + [dist_names[0][0]])
dbed.write(new_line + "\n")
dbed.close()
d = BedTool(dbed.name)
assert len(d) == len(a)
return d
def main():
p = optparse.OptionParser(__doc__)
p.add_option("-a",
dest="a",
help="file to annotate. first 3 columns are "
"chrom start stop")
p.add_option("-b", dest="b", help="superbed to annotate with")
p.add_option("--header",
dest="header",
help="a file has a header",
action="store_true",
default=False)
p.add_option("-N",
"--no-near",
dest="no_near",
help="dont find the nearest gene, just the up/downstream",
action="store_true",
default=False)
p.add_option("--upstream",
dest="upstream",
type=int,
default=None,
help="distance upstream of [a] to look for [b]")
p.add_option("--downstream",
dest="downstream",
type=int,
default=None,
help="distance downstream of [a] to look for [b]")
p.add_option("--transcripts",
dest="transcripts",
action="store_true",
default=False,
help="use transcript names in output as well as"
" gene name. default is just gene name")
p.add_option(
"--bstrand",
dest="bstrand",
default=None,
type=int,
help="if this is specified, it's the column number of the"
" strand info from the b file and nearest are reported with"
" upstream as negative relative to this column. either '+' or '-'")
opts, args = p.parse_args()
if (opts.a is None or opts.b is None):
sys.exit(not p.print_help())
if not opts.bstrand is None:
opts.bstrand -= 1
b = opts.b
if not opts.transcripts:
b = remove_transcripts(b)
if not (opts.upstream or opts.downstream):
superanno(opts.a, b, opts.header, opts.no_near, sys.stdout,
opts.bstrand)
else:
out = open(BedTool._tmp(), "w")
superanno(opts.a, b, opts.header, opts.no_near, out, opts.bstrand)
out.close()
new_header = []
out_fh = open(out.name)
new_header = [out_fh.readline().rstrip("\r\n")] if opts.header else []
for xdir in ("upstream", "downstream"):
dist = getattr(opts, xdir)
if dist is None: continue
new_out = open(BedTool._tmp(), "w")
xstream(out_fh, b, dist, xdir, new_out)
new_header.append("%s_%i" % (xdir, dist))
new_out.close()
out_fh = open(new_out.name)
if opts.header:
print "\t".join(new_header)
for line in open(out_fh.name):
sys.stdout.write(line)
def handle_coreutils_sort_kwargs(self, prog='sort', instream=None, **kwargs):
"""
Handle coreutils sort program calls.
*kwargs* are passed directly from the calling method (self.coreutils_sort).
This method figures out, given how this BedTool was constructed, what
to send to BEDTools programs -- for example, an open file to stdin with
the `-` argument, or a filename with the `-a` argument.
*instream* can be e.g., self.fn or 'a.bed' or an iterator.
"""
pybedtools.logger.debug(
'BedTool.handle_coreutils_sort_kwargs() got these kwargs:\n%s',
pprint.pformat(kwargs))
stdin = None
# Decide how to send instream to sort.
# If it's a BedTool, then get underlying stream
if isinstance(instream, BedTool):
instream = instream.fn
# Filename? No pipe, just provide the file
if isinstance(instream, six.string_types):
stdin = None
input_fn = instream
# A generator or iterator: pipe it as a generator of lines
else:
stdin = (str(i) for i in instream)
input_fn = '-'
# If stream not specified, then a tempfile will be created
if kwargs.pop('stream', None):
tmp = None
else:
output = kwargs.pop('output', None)
if output:
tmp = output
else:
tmp = BedTool._tmp()
additional_args = kwargs.pop('additional_args', None)
# Parse the kwargs into BEDTools-ready args
cmds = [prog]
for key, value in sorted(list(kwargs.items()), reverse=True):
if isinstance(value, bool):
if value:
cmds.append('--' + key)
else:
continue
elif isinstance(value, list) or isinstance(value, tuple):
value = list(map(str, value))
# sort --key 1,1 --key 2,2r -k 5,5
for val in value:
if len(key) == 1:
cmds.append('-' + key)
else:
cmds.append('--' + key)
cmds.append(str(val))
else:
cmds.append('--' + key)
cmds.append(str(value))
if additional_args:
cmds.append(additional_args)
cmds.append(input_fn)
return cmds, tmp, stdin
