def novo(args):
"""
%prog novo reads.fastq
Reference-free tGBS pipeline.
"""
from jcvi.assembly.kmer import jellyfish, histogram
from jcvi.assembly.preprocess import diginorm
from jcvi.formats.fasta import filter as fasta_filter, format
from jcvi.apps.cdhit import filter as cdhit_filter
p = OptionParser(novo.__doc__)
p.add_option("--technology", choices=("illumina", "454", "iontorrent"),
default="iontorrent", help="Sequencing platform")
p.add_option("--dedup", choices=("uclust", "cdhit"),
default="cdhit", help="Dedup algorithm")
p.set_depth(depth=50)
p.set_align(pctid=96)
p.set_home("cdhit", default="/usr/local/bin/")
p.set_home("fiona", default="/usr/local/bin/")
p.set_home("jellyfish", default="/usr/local/bin/")
p.set_cpus()
opts, args = p.parse_args(args)
if len(args) != 1:
sys.exit(not p.print_help())
fastqfile, = args
cpus = opts.cpus
depth = opts.depth
pf, sf = fastqfile.rsplit(".", 1)
diginormfile = pf + ".diginorm." + sf
if need_update(fastqfile, diginormfile):
diginorm([fastqfile, "--single", "--depth={0}".format(depth)])
keepabund = fastqfile + ".keep.abundfilt"
sh("cp -s {0} {1}".format(keepabund, diginormfile))
jf = pf + "-K23.histogram"
if need_update(diginormfile, jf):
jellyfish([diginormfile, "--prefix={0}".format(pf),
"--cpus={0}".format(cpus),
"--jellyfish_home={0}".format(opts.jellyfish_home)])
genomesize = histogram([jf, pf, "23"])
fiona = pf + ".fiona.fa"
if need_update(diginormfile, fiona):
cmd = op.join(opts.fiona_home, "fiona")
cmd += " -g {0} -nt {1} --sequencing-technology {2}".\
format(genomesize, cpus, opts.technology)
cmd += " -vv {0} {1}".format(diginormfile, fiona)
logfile = pf + ".fiona.log"
sh(cmd, outfile=logfile, errfile=logfile)
dedup = opts.dedup
pctid = opts.pctid
cons = fiona + ".P{0}.{1}.consensus.fasta".format(pctid, dedup)
if need_update(fiona, cons):
if dedup == "cdhit":
deduplicate([fiona, "--consensus", "--reads",
"--pctid={0}".format(pctid),
"--cdhit_home={0}".format(opts.cdhit_home)])
else:
uclust([fiona, "--pctid={0}".format(pctid)])
filteredfile = pf + ".filtered.fasta"
if need_update(cons, filteredfile):
covfile = pf + ".cov.fasta"
cdhit_filter([cons, "--outfile={0}".format(covfile),
"--minsize={0}".format(depth / 5)])
fasta_filter([covfile, "50", "--outfile={0}".format(filteredfile)])
finalfile = pf + ".final.fasta"
if need_update(filteredfile, finalfile):
format([filteredfile, finalfile, "--sequential=replace",
"--prefix={0}_".format(pf)])
def novo(args):
"""
%prog novo reads.fastq
Reference-free tGBS pipeline.
"""
from jcvi.assembly.kmer import jellyfish, histogram
from jcvi.assembly.preprocess import diginorm
from jcvi.formats.fasta import filter as fasta_filter, format
from jcvi.apps.cdhit import filter as cdhit_filter
p = OptionParser(novo.__doc__)
p.add_option("--technology",
choices=("illumina", "454", "iontorrent"),
default="iontorrent",
help="Sequencing platform")
p.add_option("--dedup",
choices=("uclust", "cdhit"),
default="cdhit",
help="Dedup algorithm")
p.set_depth(depth=50)
p.set_align(pctid=96)
p.set_home("cdhit")
p.set_home("fiona")
p.set_cpus()
opts, args = p.parse_args(args)
if len(args) != 1:
sys.exit(not p.print_help())
fastqfile, = args
cpus = opts.cpus
depth = opts.depth
pf, sf = fastqfile.rsplit(".", 1)
diginormfile = pf + ".diginorm." + sf
if need_update(fastqfile, diginormfile):
diginorm([fastqfile, "--single", "--depth={0}".format(depth)])
keepabund = fastqfile + ".keep.abundfilt"
sh("cp -s {0} {1}".format(keepabund, diginormfile))
jf = pf + "-K23.histogram"
if need_update(diginormfile, jf):
jellyfish([
diginormfile, "--prefix={0}".format(pf), "--cpus={0}".format(cpus)
])
genomesize = histogram([jf, pf, "23"])
fiona = pf + ".fiona.fa"
if need_update(diginormfile, fiona):
cmd = op.join(opts.fiona_home, "bin/fiona")
cmd += " -g {0} -nt {1} --sequencing-technology {2}".\
format(genomesize, cpus, opts.technology)
cmd += " -vv {0} {1}".format(diginormfile, fiona)
logfile = pf + ".fiona.log"
sh(cmd, outfile=logfile, errfile=logfile)
dedup = opts.dedup
pctid = opts.pctid
cons = fiona + ".P{0}.{1}.consensus.fasta".format(pctid, dedup)
if need_update(fiona, cons):
if dedup == "cdhit":
deduplicate([
fiona, "--consensus", "--reads", "--pctid={0}".format(pctid),
"--cdhit_home={0}".format(opts.cdhit_home)
])
else:
uclust([fiona, "--pctid={0}".format(pctid)])
filteredfile = pf + ".filtered.fasta"
if need_update(cons, filteredfile):
covfile = pf + ".cov.fasta"
cdhit_filter([
cons, "--outfile={0}".format(covfile),
"--minsize={0}".format(depth / 5)
])
fasta_filter([covfile, "50", "--outfile={0}".format(filteredfile)])
finalfile = pf + ".final.fasta"
if need_update(filteredfile, finalfile):
format([
filteredfile, finalfile, "--sequential=replace",
"--prefix={0}_".format(pf)
])