def liftover(args):
"""
%prog liftover agpfile bedfile
Given coordinates in components, convert to the coordinates in chromosomes.
"""
p = OptionParser(liftover.__doc__)
p.add_option("--prefix", default=False, action="store_true",
help="Prepend prefix to accn names [default: %default]")
opts, args = p.parse_args(args)
if len(args) != 2:
sys.exit(p.print_help())
agpfile, bedfile = args
agp = AGP(agpfile).order
bed = Bed(bedfile)
newbed = Bed()
for b in bed:
component = b.seqid
if component not in agp:
newbed.append(b)
continue
i, a = agp[component]
assert a.component_beg < a.component_end
arange = a.component_beg, a.component_end
assert b.start < b.end
brange = b.start, b.end
st = range_intersect(arange, brange)
if not st:
continue
start, end = st
assert start <= end
if a.orientation == '-':
d = a.object_end + a.component_beg
s, t = d - end, d - start
else:
d = a.object_beg - a.component_beg
s, t = d + start, d + end
name = b.accn.replace(" ", "_")
if opts.prefix:
name = component + "_" + name
bline = "\t".join(str(x) for x in (a.object, s - 1, t, name))
newbed.append(BedLine(bline))
newbed.sort(key=newbed.nullkey)
newbed.print_to_file()
def bed(args):
'''
%prog bed gff_file [--options]
Parses the start, stop locations of the selected features out of GFF and
generate a bed file
'''
p = OptionParser(bed.__doc__)
p.add_option(
"--type",
dest="type",
default="gene",
help=
"Feature type to extract, use comma for multiple [default: %default]")
p.add_option("--key",
dest="key",
default="ID",
help="Key in the attributes to extract [default: %default]")
set_outfile(p)
opts, args = p.parse_args(args)
if len(args) != 1:
sys.exit(not p.print_help())
gffile, = args
key = opts.key
if key == "None":
key = None
type = set(x.strip() for x in opts.type.split(","))
gff = Gff(gffile, key=key)
b = Bed()
for g in gff:
if g.type not in type:
continue
b.append(g.bedline)
b.sort(key=b.key)
b.print_to_file(opts.outfile)
def bed(args):
'''
%prog bed gff_file [--options]
Parses the start, stop locations of the selected features out of GFF and
generate a bed file
'''
p = OptionParser(bed.__doc__)
p.add_option("--type", dest="type", default="gene",
help="Feature type to extract, use comma for multiple [default: %default]")
p.add_option("--key", dest="key", default="ID",
help="Key in the attributes to extract [default: %default]")
set_outfile(p)
opts, args = p.parse_args(args)
if len(args) != 1:
sys.exit(not p.print_help())
gffile, = args
key = opts.key
if key == "None":
key = None
type = set(x.strip() for x in opts.type.split(","))
gff = Gff(gffile, key=key)
b = Bed()
for g in gff:
if g.type not in type:
continue
b.append(g.bedline)
b.sort(key=b.key)
b.print_to_file(opts.outfile)
def rename(args):
"""
%prog rename genes.bed [gaps.bed]
Rename genes for annotation release.
For genes on chromosomes (e.g. the 12th gene on C1):
Bo1g00120
For genes on scaffolds (e.g. the 12th gene on unplaced Scaffold00285):
Bo00285s120
The genes identifiers will increment by 10. So assuming no gap, these are
the consecutive genes:
Bo1g00120, Bo1g00130, Bo1g00140...
Bo00285s120, Bo00285s130, Bo00285s140...
When we encounter gaps, we would like the increment to be larger. For example,
Bo1g00120, <gap>, Bo1g01120...
Gaps bed file is optional.
"""
import string
p = OptionParser(rename.__doc__)
p.add_option("-a",
dest="gene_increment",
default=10,
type="int",
help="Increment for continuous genes [default: %default]")
p.add_option("-b",
dest="gap_increment",
default=1000,
type="int",
help="Increment for gaps [default: %default]")
p.add_option("--pad0",
default=6,
type="int",
help="Pad gene identifiers with 0 [default: %default]")
p.add_option(
"--spad0",
default=4,
type="int",
help="Pad gene identifiers on small scaffolds [default: %default]")
p.add_option("--prefix",
default="Bo",
help="Genome prefix [default: %default]")
p.add_option("--jgi", default=False, action="store_true",
help="Create JGI style identifier PREFIX.NN[G|TE]NNNNN.1" + \
" [default: %default]")
opts, args = p.parse_args(args)
if len(args) not in (1, 2):
sys.exit(not p.print_help())
genebed = args[0]
gapbed = args[1] if len(args) == 2 else None
prefix = opts.prefix
gene_increment = opts.gene_increment
gap_increment = opts.gap_increment
genes = Bed(genebed)
if gapbed:
fp = open(gapbed)
for row in fp:
genes.append(BedLine(row))
genes.sort(key=genes.key)
idsfile = prefix + ".ids"
newbedfile = prefix + ".bed"
gap_increment -= gene_increment
assert gap_increment >= 0
if opts.jgi:
prefix += "."
fw = open(idsfile, "w")
for chr, lines in groupby(genes, key=lambda x: x.seqid):
lines = list(lines)
pad0 = opts.pad0 if len(lines) > 1000 else opts.spad0
isChr = chr[0].upper() == 'C'
digits = "".join(x for x in chr if x in string.digits)
gs = "g" if isChr else "s"
pp = prefix + digits + gs
idx = 0
if isChr:
idx += gap_increment
for r in lines:
isGap = r.strand not in ("+", "-")
if isGap:
idx += gap_increment
continue
else:
idx += gene_increment
accn = pp + "{0:0{1}d}".format(idx, pad0)
oldaccn = r.accn
print >> fw, "\t".join((oldaccn, accn))
r.accn = accn
genes.print_to_file(newbedfile)
logging.debug("Converted IDs written to `{0}`.".format(idsfile))
logging.debug("Converted bed written to `{0}`.".format(newbedfile))
def rename(args):
"""
%prog rename genes.bed [gaps.bed]
Rename genes for annotation release.
For genes on chromosomes (e.g. the 12th gene on C1):
Bo1g00120
For genes on scaffolds (e.g. the 12th gene on unplaced Scaffold00285):
Bo00285s120
The genes identifiers will increment by 10. So assuming no gap, these are
the consecutive genes:
Bo1g00120, Bo1g00130, Bo1g00140...
Bo00285s120, Bo00285s130, Bo00285s140...
When we encounter gaps, we would like the increment to be larger. For example,
Bo1g00120, <gap>, Bo1g01120...
Gaps bed file is optional.
"""
import string
p = OptionParser(rename.__doc__)
p.add_option("-a", dest="gene_increment", default=10, type="int",
help="Increment for continuous genes [default: %default]")
p.add_option("-b", dest="gap_increment", default=1000, type="int",
help="Increment for gaps [default: %default]")
p.add_option("--pad0", default=6, type="int",
help="Pad gene identifiers with 0 [default: %default]")
p.add_option("--spad0", default=4, type="int",
help="Pad gene identifiers on small scaffolds [default: %default]")
p.add_option("--prefix", default="Bo",
help="Genome prefix [default: %default]")
p.add_option("--jgi", default=False, action="store_true",
help="Create JGI style identifier PREFIX.NN[G|TE]NNNNN.1" + \
" [default: %default]")
opts, args = p.parse_args(args)
if len(args) not in (1, 2):
sys.exit(not p.print_help())
genebed = args[0]
gapbed = args[1] if len(args) == 2 else None
prefix = opts.prefix
gene_increment = opts.gene_increment
gap_increment = opts.gap_increment
genes = Bed(genebed)
if gapbed:
fp = open(gapbed)
for row in fp:
genes.append(BedLine(row))
genes.sort(key=genes.key)
idsfile = prefix + ".ids"
newbedfile = prefix + ".bed"
gap_increment -= gene_increment
assert gap_increment >= 0
if opts.jgi:
prefix += "."
fw = open(idsfile, "w")
for chr, lines in groupby(genes, key=lambda x: x.seqid):
lines = list(lines)
pad0 = opts.pad0 if len(lines) > 1000 else opts.spad0
isChr = chr[0].upper() == 'C'
digits = "".join(x for x in chr if x in string.digits)
gs = "g" if isChr else "s"
pp = prefix + digits + gs
idx = 0
if isChr:
idx += gap_increment
for r in lines:
isGap = r.strand not in ("+", "-")
if isGap:
idx += gap_increment
continue
else:
idx += gene_increment
accn = pp + "{0:0{1}d}".format(idx, pad0)
oldaccn = r.accn
print >> fw, "\t".join((oldaccn, accn))
r.accn = accn
genes.print_to_file(newbedfile)
logging.debug("Converted IDs written to `{0}`.".format(idsfile))
logging.debug("Converted bed written to `{0}`.".format(newbedfile))
def install(args):
"""
%prog install patchers.bed patchers.fasta backbone.fasta alt.fasta
Install patches into backbone, using sequences from alternative assembly.
The patches sequences are generated via jcvi.assembly.patch.fill().
The output is a bedfile that can be converted to AGP using
jcvi.formats.agp.frombed().
"""
from jcvi.apps.base import blast
from jcvi.formats.blast import BlastSlow
from jcvi.formats.fasta import SeqIO
from jcvi.utils.iter import roundrobin
p = OptionParser(install.__doc__)
p.add_option(
"--rclip",
default=1,
type="int",
help="Pair ID is derived from rstrip N chars [default: %default]")
p.add_option(
"--maxsize",
default=1000000,
type="int",
help="Maximum size of patchers to be replaced [default: %default]")
p.add_option("--prefix",
help="Prefix of the new object [default: %default]")
p.add_option(
"--strict",
default=False,
action="store_true",
help="Only update if replacement has no gaps [default: %default]")
opts, args = p.parse_args(args)
if len(args) != 4:
sys.exit(not p.print_help())
pbed, pfasta, bbfasta, altfasta = args
Max = opts.maxsize # Max DNA size to replace gap
rclip = opts.rclip
prefix = opts.prefix
blastfile = blast([altfasta, pfasta, "--wordsize=100", "--pctid=99"])
order = Bed(pbed).order
beforebed, afterbed = "before.bed", "after.bed"
fwa = open(beforebed, "w")
fwb = open(afterbed, "w")
key1 = lambda x: x.query
key2 = lambda x: x.query[:-rclip] if rclip else key1
data = BlastSlow(blastfile)
for pe, lines in groupby(data, key=key2):
lines = list(lines)
if len(lines) != 2:
continue
a, b = lines
aquery, bquery = a.query, b.query
asubject, bsubject = a.subject, b.subject
if asubject != bsubject:
continue
astrand, bstrand = a.orientation, b.orientation
assert aquery[-1] == 'L' and bquery[-1] == 'R', str((aquery, bquery))
ai, ax = order[aquery]
bi, bx = order[bquery]
qstart, qstop = ax.start + a.qstart - 1, bx.start + b.qstop - 1
if astrand == '+' and bstrand == '+':
sstart, sstop = a.sstart, b.sstop
elif astrand == '-' and bstrand == '-':
sstart, sstop = b.sstart, a.sstop
else:
continue
if sstart > sstop:
continue
if sstop > sstart + Max:
continue
name = aquery[:-1] + "LR"
print >> fwa, "\t".join(str(x) for x in \
(ax.seqid, qstart - 1, qstop, name, 1000, "+"))
print >> fwb, "\t".join(str(x) for x in \
(asubject, sstart - 1, sstop, name, 1000, astrand))
fwa.close()
fwb.close()
beforefasta = fastaFromBed(beforebed, bbfasta, name=True, stranded=True)
afterfasta = fastaFromBed(afterbed, altfasta, name=True, stranded=True)
# Exclude the replacements that contain more Ns than before
ah = SeqIO.parse(beforefasta, "fasta")
bh = SeqIO.parse(afterfasta, "fasta")
count_Ns = lambda x: x.seq.count('n') + x.seq.count('N')
exclude = set()
for arec, brec in zip(ah, bh):
an = count_Ns(arec)
bn = count_Ns(brec)
if opts.strict:
if bn == 0:
continue
elif bn < an:
continue
id = arec.id
exclude.add(id)
logging.debug("Ignore {0} updates because of decreasing quality."\
.format(len(exclude)))
abed = Bed(beforebed, sorted=False)
bbed = Bed(afterbed, sorted=False)
abed = [x for x in abed if x.accn not in exclude]
bbed = [x for x in bbed if x.accn not in exclude]
abedfile = "before.filtered.bed"
bbedfile = "after.filtered.bed"
afbed = Bed()
afbed.extend(abed)
bfbed = Bed()
bfbed.extend(bbed)
afbed.print_to_file(abedfile)
bfbed.print_to_file(bbedfile)
# Shuffle the two bedfiles together
sz = Sizes(bbfasta)
sizes = sz.mapping
shuffled = "shuffled.bed"
border = bfbed.order
all = []
afbed.sort(key=afbed.nullkey)
totalids = len(sizes)
import math
pad = int(math.log10(totalids)) + 1
cj = 0
seen = set()
accn = lambda x: "{0}{1:0{2}d}".format(prefix, x, pad)
for seqid, aa in afbed.sub_beds():
cj += 1
abeds, bbeds, beds = [], [], []
size = sizes[seqid]
ranges = [(x.seqid, x.start, x.end) for x in aa]
cranges = range_interleave(ranges, sizes={seqid: size})
for seqid, start, end in cranges:
bedline = "\t".join(str(x) for x in (seqid, start - 1, end))
abeds.append(BedLine(bedline))
for a in aa:
gapid = a.accn
bi, b = border[gapid]
bbeds.append(b)
a = abeds[0] if abeds else []
assert abs(len(abeds) - len(bbeds)) <= 1
if (not a) or a.start > 1:
abeds, bbeds = bbeds, abeds
beds = list(roundrobin(abeds, bbeds))
if prefix:
for b in beds:
b.accn = accn(cj)
all.extend(beds)
seen.add(seqid)
# Singletons
for seqid, size in sz.iter_sizes():
if seqid in seen:
continue
bedline = "\t".join(str(x) for x in (seqid, 0, size, accn(cj)))
b = BedLine(bedline)
cj += 1
if prefix:
b.accn = accn(cj)
all.append(b)
shuffledbed = Bed()
shuffledbed.extend(all)
shuffledbed.print_to_file(shuffled)