def cnsfix(args):
"""
%prog cnsfix consensus-fix.out.FAILED > blacklist.ids
Parse consensus-fix.out to extract layouts for fixed unitigs. This will
mark all the failed fragments detected by utgcnsfix and pop them out of the
existing unitigs.
"""
from jcvi.formats.base import read_block
p = OptionParser(cnsfix.__doc__)
opts, args = p.parse_args(args)
if len(args) != 1:
sys.exit(not p.print_help())
cnsfixout, = args
fp = open(cnsfixout)
utgs = []
saves = []
for header, contents in read_block(fp, "Evaluating"):
contents = list(contents)
utg = header.split()[2]
utgs.append(utg)
# Look for this line:
# save fragment idx=388 ident=206054426 for next pass
for c in contents:
if not c.startswith("save"):
continue
ident = c.split()[3].split("=")[-1]
saves.append(ident)
print "\n".join(saves)
def dreme2tsv(args):
fh = must_open(args.fi)
start = fh.tell()
pos_c, neg_c = '', ''
for line in fh:
ps = re.split(r" +", line)
if len(ps) >= 2 and ps[1].startswith('positives'):
pos_c = ps[2]
elif len(ps) >= 2 and ps[1].startswith('negatives'):
neg_c = ps[2]
break
fh.seek(start)
i = 0
fho = must_open(args.fo, 'w')
fho.write("mid\tre\tseq\tseq_rc\tpos\tneg\tpos_c\tneg_c\tpval\teval\n")
for head, lines in read_block(fh, 'MOTIF'):
if not head: break
i += 1
mtf, conseq, mid = head.split(" ")
for line in lines:
if line.startswith("#"):
line = line.replace("#", '').strip()
ps = re.split(r" +", line)
if ps[0] in ['Word', 'Stopping', 'Running']: continue
tag = ''
if ps[0] == 'BEST':
tag = 'RE'
ps = ps[1:]
seq, seq_rc, pos, neg, pval, eval = ps
fho.write("\t".join([
mid, tag, seq, seq_rc, pos, neg, pos_c, neg_c, pval, eval
]) + '\n')
fh.close()
def __init__(self, filename):
super(ClstrFile, self).__init__(filename)
assert filename.endswith(".clstr")
fp = open(filename)
for clstr, members in read_block(fp, ">"):
self.append([ClstrLine(x) for x in members])
def __init__(self, filename):
super(ClstrFile, self).__init__(filename)
assert filename.endswith(".clstr")
fp = open(filename)
for clstr, members in read_block(fp, ">"):
self.append([ClstrLine(x) for x in members])
def __init__(self, filename=None, ctgsizes=None):
super(OO, self).__init__(filename)
if filename is None:
return
from jcvi.formats.base import read_block
fp = open(filename)
prefix = "contig_"
self.contigs = set()
for header, block in read_block(fp, ">"):
header = header[1:] # Trim the '>'
header = header.split()[0]
for b in block:
ctg, orientation = b.split()
if ctg.startswith(prefix):
ctg = ctg[len(prefix):]
assert orientation in ("BE", "EB")
strand = "+" if orientation == "BE" else "-"
ctgsize = ctgsizes[ctg]
self.add(header, ctg, ctgsize, strand)
self.contigs.add(ctg)
def parse_names(lstfile):
"""
This is the alternative format `lstfile`. In this format, there are two
sections, starting with [Sequence] and [Manuscript], respectively, then
followed by authors separated by comma.
"""
from jcvi.formats.base import read_block
fp = open(lstfile)
all_authors = []
for header, seq in read_block(fp, "["):
seq = " ".join(seq)
authors = []
for au in seq.split(","):
au = au.strip()
if not au:
continue
au = string.translate(au, None, string.digits)
#au = au.replace("-", '')
authors.append(au)
all_authors.append(authors)
out = []
for authors in all_authors:
blocks = []
for au in authors:
last, first, initials = get_name_parts(au)
suffix = ""
nameblock = NameTemplate.format(last=last, first=first,
initials=initials, suffix=suffix)
blocks.append(nameblock)
bigblock = ",\n".join(blocks)
out.append(bigblock)
return out
def __init__(self, filename=None, ctgsizes=None):
super(OO, self).__init__(filename)
if filename is None:
return
from jcvi.formats.base import read_block
fp = open(filename)
prefix = "contig_"
self.contigs = set()
for header, block in read_block(fp, ">"):
header = header[1:] # Trim the '>'
header = header.split()[0]
for b in block:
ctg, orientation = b.split()
if ctg.startswith(prefix):
ctg = ctg[len(prefix):]
assert orientation in ("BE", "EB")
strand = "+" if orientation == "BE" else "-"
ctgsize = ctgsizes[ctg]
self.add(header, ctg, ctgsize, strand)
self.contigs.add(ctg)
def dust2bed(args):
"""
%prog dust2bed fastafile
Use dustmasker to find low-complexity regions (LCRs) in the genome.
"""
from jcvi.formats.base import read_block
p = OptionParser(dust2bed.__doc__)
opts, args = p.parse_args(args)
if len(args) != 1:
sys.exit(not p.print_help())
fastafile, = args
interval = fastafile + ".iv"
if need_update(fastafile, interval):
cmd = "dustmasker -in {0}".format(fastafile)
sh(cmd, outfile=interval)
fp = open(interval)
bedfile = fastafile.rsplit(".", 1)[0] + ".dust.bed"
fw = must_open(bedfile, "w")
nlines = 0
nbases = 0
for header, block in read_block(fp, ">"):
header = header.strip(">")
for b in block:
start, end = b.split(" - ")
start, end = int(start), int(end)
print >> fw, "\t".join(str(x) for x in (header, start, end))
nlines += 1
nbases += end - start
logging.debug("A total of {0} DUST intervals ({1} bp) exported to `{2}`".\
format(nlines, nbases, bedfile))
def add_score(args):
mtfs = read_streme_xml(args.fx)
fhi = open(args.fi, 'r')
fho = open(args.fo, 'w')
#
start = "MOTIF"
while 1:
pos = fhi.tell()
line = fhi.readline()
if not line:
break
if line.startswith(start):
fhi.seek(pos)
break
else:
fho.write(line)
#
i = 0
for head, content in read_block(fhi, 'MOTIF'):
pre, mid, alt = head.split(' ')
assert mtfs[i]['id'] == mid, 'motifs not in sync'
new_head = f"{pre} {mid} {mtfs[i]['score_threshold']}"
fho.write(new_head + "\n")
for line in content:
if line.startswith("0") or line.startswith("1"):
fho.write(" " + line + "\n")
else:
fho.write(line + "\n")
i += 1
fhi.close()
fho.close()
def cnsfix(args):
"""
%prog cnsfix consensus-fix.out.FAILED > blacklist.ids
Parse consensus-fix.out to extract layouts for fixed unitigs. This will
mark all the failed fragments detected by utgcnsfix and pop them out of the
existing unitigs.
"""
from jcvi.formats.base import read_block
p = OptionParser(cnsfix.__doc__)
opts, args = p.parse_args(args)
if len(args) != 1:
sys.exit(not p.print_help())
cnsfixout, = args
fp = open(cnsfixout)
utgs = []
saves = []
for header, contents in read_block(fp, "Evaluating"):
contents = list(contents)
utg = header.split()[2]
utgs.append(utg)
# Look for this line:
# save fragment idx=388 ident=206054426 for next pass
for c in contents:
if not c.startswith("save"):
continue
ident = c.split()[3].split("=")[-1]
saves.append(ident)
print "\n".join(saves)
def iter_scaffold(self):
filename = self.filename
sizes = self.sizes
fp = open(filename)
for header, lines in read_block(fp, ">"):
scaffold, size, tigs = header[1:].split("|")
lines = [EvidenceLine(x, sizes) for x in lines if x.strip()]
yield scaffold, lines
def iter_scaffold(self):
filename = self.filename
sizes = self.sizes
fp = open(filename)
for header, lines in read_block(fp, ">"):
scaffold, size, tigs = header[1:].split("|")
lines = [EvidenceLine(x, sizes) for x in lines if x.strip()]
yield scaffold, lines
def iter_chain(self):
fp = open(self.filename)
for row in fp:
if row[0] != '#':
break
for chain, lines in read_block(fp, "chain"):
lines = list(lines)
yield ChainLine(chain, lines)
def iter_chain(self):
fp = open(self.filename)
for row in fp:
if row[0] != '#':
break
for chain, lines in read_block(fp, "chain"):
lines = list(lines)
yield ChainLine(chain, lines)
def iter_records(self):
c = None
for a, b in read_block(self.fp, "#"):
if a[:2] == '##':
if c:
yield c
c = ContigLine(a)
else:
c.reads.append(ReadLine(a, c.id))
if c: # last one
yield c
def iter_records(self):
c = None
for a, b in read_block(self.fp, "#"):
if a[:2] == '##':
if c:
yield c
c = ContigLine(a)
else:
c.reads.append(ReadLine(a, c.id))
if c: # last one
yield c
def __init__(self, filename):
super(BinMap, self).__init__(filename)
fp = open(filename)
for header, seq in read_block(fp, "group "):
lg = header.split()[-1]
self[lg] = []
for s in seq:
if s.strip() == '' or s[0] == ';':
continue
marker, pos = s.split()
pos = int(float(pos) * 1000)
self[lg].append((marker, pos))
def read_trees(tree):
from urlparse import parse_qs
from jcvi.formats.base import read_block
trees = []
fp = open(tree)
for header, tx in read_block(fp, "#"):
header = parse_qs(header[1:])
label = header["label"][0].strip("\"")
outgroup = header["outgroup"]
trees.append((label, outgroup, "".join(tx)))
return trees
def read_trees(tree):
from urlparse import parse_qs
from jcvi.formats.base import read_block
trees = []
fp = open(tree)
for header, tx in read_block(fp, "#"):
header = parse_qs(header[1:])
label = header["label"][0].strip("\"")
outgroup = header["outgroup"]
trees.append((label, outgroup, "".join(tx)))
return trees
def read_trees(tree):
from six.moves.urllib.parse import parse_qs
from jcvi.formats.base import read_block
trees = []
fp = open(tree)
for header, tx in read_block(fp, "#"):
header = parse_qs(header[1:])
label = header["label"][0].strip("\"")
outgroup = header["outgroup"]
color, = header.get("color", ["k"])
trees.append((label, outgroup, color, "".join(tx)))
return trees
def read_meme(fi):
mtfs = []
fhi = open(fi, 'r')
for head, content in read_block(fhi, 'MOTIF'):
ps = head.split(' ')
pre, mid = ps[:2]
score = ''
if len(ps) >= 3:
score = ps[2]
#mtf = mid.split("-")[1]
if is_number(score):
score = float(score)
width = len(content) - 2
mtfs.append([mid, width, score])
#print(mid,'\t',width)
return mtfs
def __iter__(self):
line = self._handle.readline()
if not line.startswith(bac_tag):
read_until(self._handle, bac_tag)
for header, seq in read_block(self._handle, bac_tag):
rec = FpcRecord()
assert header.startswith(bac_tag)
rec.bac_name = header.split('\"')[1]
for line in seq:
if line.startswith("Map"):
rec.ctg_name = line.split('\"')[1]
if "Left" in line:
rec.map_left = line.split("Left")[1].split()[0]
rec.map_left = int(float(rec.map_left))
if "Right" in line:
rec.map_right = line.split("Right")[1].split()[0]
rec.map_right = int(float(rec.map_right))
if line.startswith("Gel_number"):
rec.gel_number = line.split()[-1]
if line.startswith("Fp_number"):
rec.fp_number = line.split()[-1]
if line.startswith("Bands"):
rec.bands = line.split()[-1]
if line.startswith("match_to_cosmid"):
rec.cosmid = line.split('\"')[1]
rec.cosmid_match = line.split("_match_")[0]
if line.startswith("Positive"):
rec.probes.append(line.split('\"')[1])
if line.startswith("Fpc_remark"):
rec.remark.append(line.split('\"')[1])
if line.startswith("Creation_date"):
rec.cre_date = line.split('_date')[1].strip()
if line.startswith("Modified_date"):
rec.mod_date = line.split('_date')[1].strip()
yield rec
def __iter__(self):
line = self._handle.readline()
if not line.startswith(bac_tag):
read_until(self._handle, bac_tag)
for header, seq in read_block(self._handle, bac_tag):
rec = FpcRecord()
assert header.startswith(bac_tag)
rec.bac_name = header.split('\"')[1]
for line in seq:
if line.startswith("Map"):
rec.ctg_name = line.split('\"')[1]
if "Left" in line:
rec.map_left = line.split("Left")[1].split()[0]
rec.map_left = int(float(rec.map_left))
if "Right" in line:
rec.map_right = line.split("Right")[1].split()[0]
rec.map_right = int(float(rec.map_right))
if line.startswith("Gel_number"):
rec.gel_number = line.split()[-1]
if line.startswith("Fp_number"):
rec.fp_number = line.split()[-1]
if line.startswith("Bands"):
rec.bands = line.split()[-1]
if line.startswith("match_to_cosmid"):
rec.cosmid = line.split('\"')[1]
rec.cosmid_match = line.split("_match_")[0]
if line.startswith("Positive"):
rec.probes.append(line.split('\"')[1])
if line.startswith("Fpc_remark"):
rec.remark.append(line.split('\"')[1])
if line.startswith("Creation_date"):
rec.cre_date = line.split('_date')[1].strip()
if line.startswith("Modified_date"):
rec.mod_date = line.split('_date')[1].strip()
yield rec
def coge(args):
"""
%prog coge cogefile
Convert CoGe file to anchors file.
"""
p = OptionParser(coge.__doc__)
opts, args = p.parse_args(args)
if len(args) != 1:
sys.exit(not p.print_help())
cogefile, = args
fp = must_open(cogefile)
cogefile = cogefile.replace(".gz", "")
ksfile = cogefile + ".ks"
anchorsfile = cogefile + ".anchors"
fw_ks = must_open(ksfile, "w")
fw_ac = must_open(anchorsfile, "w")
tag = "###"
print >> fw_ks, tag
for header, lines in read_block(fp, tag):
print >> fw_ac, tag
lines = list(lines)
for line in lines:
if line[0] == '#':
continue
ks, ka, achr, a, astart, astop, bchr, \
b, bstart, bstop, ev, ss = line.split()
a = a.split("||")[3]
b = b.split("||")[3]
print >> fw_ac, "\t".join((a, b, ev))
print >> fw_ks, ",".join((";".join((a, b)), ks, ka, ks, ka))
fw_ks.close()
fw_ac.close()
def coge(args):
"""
%prog coge cogefile
Convert CoGe file to anchors file.
"""
p = OptionParser(coge.__doc__)
opts, args = p.parse_args(args)
if len(args) != 1:
sys.exit(not p.print_help())
cogefile, = args
fp = must_open(cogefile)
cogefile = cogefile.replace(".gz", "")
ksfile = cogefile + ".ks"
anchorsfile = cogefile + ".anchors"
fw_ks = must_open(ksfile, "w")
fw_ac = must_open(anchorsfile, "w")
tag = "###"
print >> fw_ks, tag
for header, lines in read_block(fp, tag):
print >> fw_ac, tag
lines = list(lines)
for line in lines:
if line[0] == '#':
continue
ks, ka, achr, a, astart, astop, bchr, \
b, bstart, bstop, ev, ss = line.split()
a = a.split("||")[3]
b = b.split("||")[3]
print >> fw_ac, "\t".join((a, b, ev))
print >> fw_ks, ",".join((";".join((a, b)), ks, ka, ks, ka))
fw_ks.close()
fw_ac.close()
def summary(args):
"""
%prog summary cdhit.clstr
Parse cdhit.clstr file to get distribution of cluster sizes.
"""
from jcvi.graphics.histogram import loghistogram
p = OptionParser(summary.__doc__)
opts, args = p.parse_args(args)
if len(args) != 1:
sys.exit(not p.print_help())
clstrfile, = args
assert clstrfile.endswith(".clstr")
fp = open(clstrfile)
data = []
for clstr, members in read_block(fp, ">"):
size = len(members)
data.append(size)
loghistogram(data)
def summary(args):
"""
%prog summary cdhit.clstr
Parse cdhit.clstr file to get distribution of cluster sizes.
"""
from jcvi.graphics.histogram import loghistogram
p = OptionParser(summary.__doc__)
opts, args = p.parse_args(args)
if len(args) != 1:
sys.exit(not p.print_help())
clstrfile, = args
assert clstrfile.endswith(".clstr")
fp = open(clstrfile)
data = []
for clstr, members in read_block(fp, ">"):
size = len(members)
data.append(size)
loghistogram(data)
def iter_chain(self):
fp = open(self.filename)
for chain, lines in read_block(fp, "chain"):
lines = list(lines)
yield ChainLine(chain, lines)
def iter_blocks(self, minsize=0):
fp = open(self.filename)
for header, lines in read_block(fp, "#"):
lines = [x.split() for x in lines]
if len(lines) >= minsize:
yield lines
def iter_blocks(self, minsize=0):
fp = open(self.filename)
for header, lines in read_block(fp, "#"):
lines = [x.split() for x in lines]
if len(lines) >= minsize:
yield lines
