def extract_ends(rec, sites, flank, fw, maxfragsize=800):
"""
Extraction of ends of fragments above certain size.
"""
nsites = len(sites)
size = len(rec)
for i, s in enumerate(sites):
newid = "{0}:{1}".format(rec.name, s)
recs = []
if i == 0 or s - sites[i - 1] <= maxfragsize:
newidL = newid + "L"
left = max(s - flank, 0)
right = s
frag = rec.seq[left:right].strip("Nn")
recL = SeqRecord(frag, id=newidL, description="")
if i == 0 and s > maxfragsize: # Contig L-end
pass
else:
recs.append(recL)
if i == nsites - 1 or sites[i + 1] - s <= maxfragsize:
newidR = newid + "R"
left = s
right = min(s + flank, size)
frag = rec.seq[left:right].strip("Nn")
recR = SeqRecord(frag, id=newidR, description="")
if i == nsites - 1 and size - s > maxfragsize: # Contig R-end
pass
else:
recs.append(recR)
SeqIO.write(recs, fw, "fasta")
def emitFragment(fw,
fragID,
libID,
shredded_seq,
clr=None,
qvchar="l",
fasta=False):
"""
Print out the shredded sequence.
"""
if fasta:
s = SeqRecord(shredded_seq, id=fragID, description="")
SeqIO.write([s], fw, "fasta")
return
seq = str(shredded_seq)
slen = len(seq)
qvs = qvchar * slen # shredded reads have default low qv
if clr is None:
clr_beg, clr_end = 0, slen
else:
clr_beg, clr_end = clr
print(
frgTemplate.format(
fragID=fragID,
libID=libID,
seq=seq,
qvs=qvs,
clr_beg=clr_beg,
clr_end=clr_end,
),
file=fw,
)
def extract_full(rec, sites, flank, fw):
"""
Full extraction of seq flanking the sites.
"""
for s in sites:
newid = "{0}:{1}".format(rec.name, s)
left = max(s - flank, 0)
right = min(s + flank, len(rec))
frag = rec.seq[left:right].strip("Nn")
newrec = SeqRecord(frag, id=newid, description="")
SeqIO.write([newrec], fw, "fasta")
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 emitFragment(fw, fragID, libID, shredded_seq, fasta=False):
"""
Print out the shredded sequence.
"""
if fasta:
s = SeqRecord(shredded_seq, id=fragID, description="")
SeqIO.write([s], fw, "fasta")
return
seq = str(shredded_seq)
slen = len(seq)
qvs = DEFAULTQV * slen # shredded reads have default low qv
print >> fw, frgTemplate.format(fragID=fragID, libID=libID,
seq=fill(seq), qvs=fill(qvs), slen=slen)
def load(args):
'''
%prog load gff_file fasta_file [--options]
Parses the selected features out of GFF, with subfeatures concatenated.
For example, to get the CDS sequences, do this::
$ %prog load athaliana.gff athaliana.fa --parents mRNA --children CDS
'''
from jcvi.formats.fasta import Seq, SeqRecord
p = OptionParser(load.__doc__)
p.add_option(
"--parents",
dest="parents",
default="mRNA",
help="list of features to extract, use comma to separate (e.g."
"'gene,mRNA') [default: %default]")
p.add_option(
"--children",
dest="children",
default="CDS",
help="list of features to extract, use comma to separate (e.g."
"'five_prime_UTR,CDS,three_prime_UTR') [default: %default]")
p.add_option("--attribute",
help="The attribute field to extract [default: %default]")
set_outfile(p)
opts, args = p.parse_args(args)
if len(args) != 2:
sys.exit(p.print_help())
gff_file, fasta_file = args
g = make_index(gff_file)
f = Fasta(fasta_file, index=False)
fw = must_open(opts.outfile, "w")
parents = set(opts.parents.split(','))
children_list = set(opts.children.split(','))
attr = opts.attribute
for feat in get_parents(gff_file, parents):
children = []
for c in g.children(feat.id, 1):
if c.featuretype not in children_list:
continue
child = f.sequence(
dict(chr=c.chrom, start=c.start, stop=c.stop, strand=c.strand))
children.append((child, c))
if not children:
print >>sys.stderr, "[warning] %s has no children with type %s" \
% (feat.id, ','.join(children_list))
continue
# sort children in incremental position
children.sort(key=lambda x: x[1].start)
# reverse children if negative strand
if feat.strand == '-':
children.reverse()
feat_seq = ''.join(x[0] for x in children)
description = ",".join(feat.attributes[attr]) \
if attr and attr in feat.attributes else ""
description = description.replace("\"", "")
rec = SeqRecord(Seq(feat_seq), id=feat.id, description=description)
SeqIO.write([rec], fw, "fasta")
fw.flush()
