def fpkm(args):
"""
%prog fpkm fastafile *.bam
Calculate FPKM values from BAM file.
"""
p = OptionParser(fpkm.__doc__)
opts, args = p.parse_args(args)
if len(args) < 2:
sys.exit(not p.print_help())
fastafile = args[0]
bamfiles = args[1:]
# Create a DUMMY gff file for cuffdiff
gffile = fastafile.rsplit(".", 1)[0] + ".gff"
if need_update(fastafile, gffile):
fw = open(gffile, "w")
f = Fasta(fastafile, lazy=True)
for key, size in f.itersizes_ordered():
print >> fw, "\t".join(str(x) for x in (key, "dummy", "transcript",\
1, size, ".", ".", ".", "ID=" + key))
fw.close()
logging.debug("Dummy GFF created: {0}".format(gffile))
cmd = "cuffdiff {0} {1}".format(gffile, " ".join(bamfiles))
sh(cmd)
def __init__(self, filename, select=None):
assert op.exists(filename), "File `{0}` not found".format(filename)
# filename can be both .sizes file or FASTA formatted file
sizesname = filename
if not filename.endswith(".sizes"):
sizesname = filename + ".sizes"
filename = get_abs_path(filename)
if need_update(filename, sizesname):
cmd = "faSize"
if which(cmd):
cmd += " -detailed {0}".format(filename)
sh(cmd, outfile=sizesname)
else:
from jcvi.formats.fasta import Fasta
f = Fasta(filename)
fw = open(sizesname, "w")
for k, size in f.itersizes_ordered():
print >> fw, "\t".join((k, str(size)))
fw.close()
filename = sizesname
assert filename.endswith(".sizes")
super(Sizes, self).__init__(filename)
self.fp = open(filename)
self.filename = filename
# get sizes for individual contigs, both in list and dict
# this is to preserve the input order in the sizes file
sizes = list(self.iter_sizes())
if select:
assert select > 0
sizes = [x for x in sizes if x[1] >= select]
self.sizes_mapping = dict(sizes)
# get cumulative sizes, both in list and dict
ctgs, sizes = zip(*sizes)
self.sizes = sizes
cumsizes = np.cumsum([0] + list(sizes))
self.ctgs = ctgs
self.cumsizes = cumsizes
self.cumsizes_mapping = dict(zip(ctgs, cumsizes))
def fpkm(args):
"""
%prog fpkm fastafile *.bam
Calculate FPKM values from BAM file.
"""
p = OptionParser(fpkm.__doc__)
opts, args = p.parse_args(args)
if len(args) < 2:
sys.exit(not p.print_help())
fastafile = args[0]
bamfiles = args[1:]
# Create a DUMMY gff file for cuffdiff
gffile = fastafile.rsplit(".", 1)[0] + ".gff"
if need_update(fastafile, gffile):
fw = open(gffile, "w")
f = Fasta(fastafile, lazy=True)
for key, size in f.itersizes_ordered():
print(
"\t".join(
str(x) for x in (
key,
"dummy",
"transcript",
1,
size,
".",
".",
".",
"ID=" + key,
)),
file=fw,
)
fw.close()
logging.debug("Dummy GFF created: {0}".format(gffile))
cmd = "cuffdiff {0} {1}".format(gffile, " ".join(bamfiles))
sh(cmd)
def fasta(args):
"""
%prog fasta fastafile
Convert reads formatted as FASTA file, and convert to CA frg file. If .qual
file is found, then use it, otherwise just make a fake qual file. Mates are
assumed as adjacent sequence records (i.e. /1, /2, /1, /2 ...) unless a
matefile is given.
"""
from jcvi.formats.fasta import clean, make_qual
p = OptionParser(fasta.__doc__)
p.add_option("--clean", default=False, action="store_true", help="Clean up irregular chars in seq")
p.add_option("--matefile", help="Matepairs file")
p.add_option("--maxreadlen", default=262143, type="int", help="Maximum read length allowed")
p.add_option("--minreadlen", default=1000, type="int", help="Minimum read length allowed")
p.add_option("--sequential", default=False, action="store_true", help="Overwrite read name (e.g. long Pacbio name)")
p.set_size()
opts, args = p.parse_args(args)
if len(args) != 1:
sys.exit(not p.print_help())
fastafile, = args
maxreadlen = opts.maxreadlen
minreadlen = opts.minreadlen
if maxreadlen > 0:
split = False
f = Fasta(fastafile, lazy=True)
for id, size in f.itersizes_ordered():
if size > maxreadlen:
logging.debug("Sequence {0} (size={1}) longer than max read len {2}".format(id, size, maxreadlen))
split = True
break
if split:
for f in split_fastafile(fastafile, maxreadlen=maxreadlen):
fasta([f, "--maxreadlen=0"])
return
plate = op.basename(fastafile).split(".")[0]
mated = opts.size != 0
mean, sv = get_mean_sv(opts.size)
if mated:
libname = "Sanger{0}Kb-".format(opts.size / 1000) + plate
else:
libname = plate
frgfile = libname + ".frg"
if opts.clean:
cleanfasta = fastafile.rsplit(".", 1)[0] + ".clean.fasta"
if need_update(fastafile, cleanfasta):
clean([fastafile, "--canonical", "-o", cleanfasta])
fastafile = cleanfasta
if mated:
qualfile = make_qual(fastafile, score=21)
if opts.matefile:
matefile = opts.matefile
assert op.exists(matefile)
else:
matefile = make_matepairs(fastafile)
cmd = "convert-fasta-to-v2.pl"
cmd += " -l {0} -s {1} -q {2} ".format(libname, fastafile, qualfile)
if mated:
cmd += "-mean {0} -stddev {1} -m {2} ".format(mean, sv, matefile)
sh(cmd, outfile=frgfile)
return
fw = must_open(frgfile, "w")
print >> fw, headerTemplate.format(libID=libname)
sequential = opts.sequential
i = j = 0
for fragID, seq in parse_fasta(fastafile):
if len(seq) < minreadlen:
j += 1
continue
i += 1
if sequential:
fragID = libname + str(100000000 + i)
emitFragment(fw, fragID, libname, seq)
fw.close()
logging.debug("A total of {0} fragments written to `{1}` ({2} discarded).".format(i, frgfile, j))
def fasta(args):
"""
%prog fasta fastafile
Convert reads formatted as FASTA file, and convert to CA frg file. If .qual
file is found, then use it, otherwise just make a fake qual file. Mates are
assumed as adjacent sequence records (i.e. /1, /2, /1, /2 ...) unless a
matefile is given.
"""
from jcvi.formats.fasta import clean, make_qual
p = OptionParser(fasta.__doc__)
p.add_option("-m", dest="matefile", default=None, help="matepairs file")
p.add_option("--maxreadlen",
default=32000,
type="int",
help="Maximum read length allowed [default: %default]")
p.set_size()
opts, args = p.parse_args(args)
if len(args) != 1:
sys.exit(not p.print_help())
fastafile, = args
maxreadlen = opts.maxreadlen
f = Fasta(fastafile, lazy=True)
if maxreadlen > 0:
split = False
for id, size in f.itersizes_ordered():
if size > maxreadlen:
logging.debug("Sequence {0} (size={1}) longer than max read len {2}".\
format(id, size, maxreadlen))
split = True
break
if split:
for f in split_fastafile(fastafile, maxreadlen=maxreadlen):
fasta([f, "--maxreadlen=0"])
return
plate = op.basename(fastafile).split(".")[0]
mated = (opts.size != 0)
mean, sv = get_mean_sv(opts.size)
if mated:
libname = "Sanger{0}Kb-".format(opts.size / 1000) + plate
else:
libname = "SangerFrags-" + plate
frgfile = libname + ".frg"
cleanfasta = fastafile.rsplit(".", 1)[0] + ".clean.fasta"
if need_update(fastafile, cleanfasta):
clean([fastafile, "--canonical", "-o", cleanfasta])
fastafile = cleanfasta
qualfile = make_qual(fastafile, score=21)
if mated:
if opts.matefile:
matefile = opts.matefile
assert op.exists(matefile)
else:
matefile = make_matepairs(fastafile)
cmd = "convert-fasta-to-v2.pl"
cmd += " -l {0} -s {1} -q {2} ".\
format(libname, fastafile, qualfile)
if mated:
cmd += "-mean {0} -stddev {1} -m {2} ".format(mean, sv, matefile)
sh(cmd, outfile=frgfile)
def fasta(args):
"""
%prog fasta fastafile
Convert reads formatted as FASTA file, and convert to CA frg file. If .qual
file is found, then use it, otherwise just make a fake qual file. Mates are
assumed as adjacent sequence records (i.e. /1, /2, /1, /2 ...) unless a
matefile is given.
"""
from jcvi.formats.fasta import clean, make_qual
p = OptionParser(fasta.__doc__)
p.add_option(
"--clean",
default=False,
action="store_true",
help="Clean up irregular chars in seq",
)
p.add_option("--matefile", help="Matepairs file")
p.add_option("--maxreadlen",
default=262143,
type="int",
help="Maximum read length allowed")
p.add_option("--minreadlen",
default=1000,
type="int",
help="Minimum read length allowed")
p.add_option(
"--sequential",
default=False,
action="store_true",
help="Overwrite read name (e.g. long Pacbio name)",
)
p.set_size()
opts, args = p.parse_args(args)
if len(args) != 1:
sys.exit(not p.print_help())
(fastafile, ) = args
maxreadlen = opts.maxreadlen
minreadlen = opts.minreadlen
if maxreadlen > 0:
split = False
f = Fasta(fastafile, lazy=True)
for id, size in f.itersizes_ordered():
if size > maxreadlen:
logging.debug(
"Sequence {0} (size={1}) longer than max read len {2}".
format(id, size, maxreadlen))
split = True
break
if split:
for f in split_fastafile(fastafile, maxreadlen=maxreadlen):
fasta([f, "--maxreadlen=0"])
return
plate = op.basename(fastafile).split(".")[0]
mated = opts.size != 0
mean, sv = get_mean_sv(opts.size)
if mated:
libname = "Sanger{0}Kb-".format(opts.size / 1000) + plate
else:
libname = plate
frgfile = libname + ".frg"
if opts.clean:
cleanfasta = fastafile.rsplit(".", 1)[0] + ".clean.fasta"
if need_update(fastafile, cleanfasta):
clean([fastafile, "--canonical", "-o", cleanfasta])
fastafile = cleanfasta
if mated:
qualfile = make_qual(fastafile, score=21)
if opts.matefile:
matefile = opts.matefile
assert op.exists(matefile)
else:
matefile = make_matepairs(fastafile)
cmd = "convert-fasta-to-v2.pl"
cmd += " -l {0} -s {1} -q {2} ".format(libname, fastafile, qualfile)
if mated:
cmd += "-mean {0} -stddev {1} -m {2} ".format(mean, sv, matefile)
sh(cmd, outfile=frgfile)
return
fw = must_open(frgfile, "w")
print(headerTemplate.format(libID=libname), file=fw)
sequential = opts.sequential
i = j = 0
for fragID, seq in parse_fasta(fastafile):
if len(seq) < minreadlen:
j += 1
continue
i += 1
if sequential:
fragID = libname + str(100000000 + i)
emitFragment(fw, fragID, libname, seq)
fw.close()
logging.debug(
"A total of {0} fragments written to `{1}` ({2} discarded).".format(
i, frgfile, j))
def fasta(args):
"""
%prog fasta fastafile
Convert reads formatted as FASTA file, and convert to CA frg file. If .qual
file is found, then use it, otherwise just make a fake qual file. Mates are
assumed as adjacent sequence records (i.e. /1, /2, /1, /2 ...) unless a
matefile is given.
"""
from jcvi.formats.fasta import clean, make_qual
p = OptionParser(fasta.__doc__)
p.add_option("-m", dest="matefile", default=None,
help="matepairs file")
p.add_option("--maxreadlen", default=32000, type="int",
help="Maximum read length allowed [default: %default]")
p.set_size()
opts, args = p.parse_args(args)
if len(args) != 1:
sys.exit(not p.print_help())
fastafile, = args
maxreadlen = opts.maxreadlen
f = Fasta(fastafile, lazy=True)
if maxreadlen > 0:
split = False
for id, size in f.itersizes_ordered():
if size > maxreadlen:
logging.debug("Sequence {0} (size={1}) longer than max read len {2}".\
format(id, size, maxreadlen))
split = True
break
if split:
for f in split_fastafile(fastafile, maxreadlen=maxreadlen):
fasta([f, "--maxreadlen=0"])
return
plate = op.basename(fastafile).split(".")[0]
mated = (opts.size != 0)
mean, sv = get_mean_sv(opts.size)
if mated:
libname = "Sanger{0}Kb-".format(opts.size / 1000) + plate
else:
libname = "SangerFrags-" + plate
frgfile = libname + ".frg"
cleanfasta = fastafile.rsplit(".", 1)[0] + ".clean.fasta"
if need_update(fastafile, cleanfasta):
clean([fastafile, "--canonical", "-o", cleanfasta])
fastafile = cleanfasta
qualfile = make_qual(fastafile, score=21)
if mated:
if opts.matefile:
matefile = opts.matefile
assert op.exists(matefile)
else:
matefile = make_matepairs(fastafile)
cmd = "convert-fasta-to-v2.pl"
cmd += " -l {0} -s {1} -q {2} ".\
format(libname, fastafile, qualfile)
if mated:
cmd += "-mean {0} -stddev {1} -m {2} ".format(mean, sv, matefile)
sh(cmd, outfile=frgfile)
