def overlapbatch(args):
"""
%prog overlapbatch ctgfasta poolfasta
Fish out the sequences in `poolfasta` that overlap with `ctgfasta`.
Mix and combine using `minimus2`.
"""
p = OptionParser(overlap.__doc__)
opts, args = p.parse_args(args)
if len(args) != 2:
sys.exit(not p.print_help())
ctgfasta, poolfasta = args
f = Fasta(ctgfasta)
for k, rec in f.iteritems_ordered():
fastafile = k + ".fasta"
fw = open(fastafile, "w")
SeqIO.write([rec], fw, "fasta")
fw.close()
overlap([fastafile, poolfasta])
def overlapbatch(args):
"""
%prog overlapbatch ctgfasta poolfasta
Fish out the sequences in `poolfasta` that overlap with `ctgfasta`.
Mix and combine using `minimus2`.
"""
p = OptionParser(overlap.__doc__)
opts, args = p.parse_args(args)
if len(args) != 2:
sys.exit(not p.print_help())
ctgfasta, poolfasta = args
f = Fasta(ctgfasta)
for k, rec in f.iteritems_ordered():
fastafile = k + ".fasta"
fw = open(fastafile, "w")
SeqIO.write([rec], fw, "fasta")
fw.close()
overlap([fastafile, poolfasta])
def circular(args):
"""
%prog circular fastafile startpos
Make circular genome, startpos is the place to start the sequence. This can
be determined by mapping to a reference. Self overlaps are then resolved.
Startpos is 1-based.
"""
from jcvi.assembly.goldenpath import overlap
p = OptionParser(circular.__doc__)
p.add_option(
"--flip",
default=False,
action="store_true",
help="Reverse complement the sequence",
)
p.set_outfile()
opts, args = p.parse_args(args)
if len(args) != 2:
sys.exit(not p.print_help())
fastafile, startpos = args
startpos = int(startpos)
key, seq = next(parse_fasta(fastafile))
aseq = seq[startpos:]
bseq = seq[:startpos]
aseqfile, bseqfile = "a.seq", "b.seq"
for f, s in zip((aseqfile, bseqfile), (aseq, bseq)):
fw = must_open(f, "w")
print(">{0}\n{1}".format(f, s), file=fw)
fw.close()
o = overlap([aseqfile, bseqfile])
seq = aseq[:o.qstop] + bseq[o.sstop:]
seq = Seq(seq)
if opts.flip:
seq = seq.reverse_complement()
for f in (aseqfile, bseqfile):
os.remove(f)
fw = must_open(opts.outfile, "w")
rec = SeqRecord(seq, id=key, description="")
SeqIO.write([rec], fw, "fasta")
fw.close()
def circular(args):
"""
%prog circular fastafile startpos
Make circular genome, startpos is the place to start the sequence. This can
be determined by mapping to a reference. Self overlaps are then resolved.
Startpos is 1-based.
"""
from jcvi.assembly.goldenpath import overlap
p = OptionParser(circular.__doc__)
p.add_option("--flip", default=False, action="store_true",
help="Reverse complement the sequence")
p.set_outfile()
opts, args = p.parse_args(args)
if len(args) != 2:
sys.exit(not p.print_help())
fastafile, startpos = args
startpos = int(startpos)
key, seq = parse_fasta(fastafile).next()
aseq = seq[startpos:]
bseq = seq[:startpos]
aseqfile, bseqfile = "a.seq", "b.seq"
for f, s in zip((aseqfile, bseqfile), (aseq, bseq)):
fw = must_open(f, "w")
print >> fw, ">{0}\n{1}".format(f, s)
fw.close()
o = overlap([aseqfile, bseqfile])
seq = aseq[:o.qstop] + bseq[o.sstop:]
seq = Seq(seq)
if opts.flip:
seq = seq.reverse_complement()
for f in (aseqfile, bseqfile):
os.remove(f)
fw = must_open(opts.outfile, "w")
rec = SeqRecord(seq, id=key, description="")
SeqIO.write([rec], fw, "fasta")
fw.close()
