def align(args):
"""
%prog align database.fasta read1.fq read2.fq
Wrapper for `gsnap` single-end or paired-end, depending on the number of
args.
"""
from jcvi.formats.fasta import join
from jcvi.formats.fastq import guessoffset
from jcvi.projects.tgbs import snp
p = OptionParser(align.__doc__)
p.add_option("--join", default=False, action="store_true",
help="Join sequences with padded 50Ns")
p.add_option("--rnaseq", default=False, action="store_true",
help="Input is RNA-seq reads, turn splicing on")
p.add_option("--snp", default=False, action="store_true",
help="Call SNPs after GSNAP")
p.set_cpus()
opts, args = p.parse_args(args)
if len(args) == 2:
logging.debug("Single-end alignment")
elif len(args) == 3:
logging.debug("Paired-end alignment")
else:
sys.exit(not p.print_help())
dbfile, readfile = args[0:2]
if opts.join:
dbfile = join([dbfile, "--gapsize=50", "--newid=chr1"])
assert op.exists(dbfile) and op.exists(readfile)
prefix = get_prefix(readfile, dbfile)
logfile = prefix + ".log"
gsnapfile = prefix + ".gsnap"
if not need_update((dbfile, readfile), gsnapfile):
logging.error("`{0}` exists. `gsnap` already run.".format(gsnapfile))
else:
dbdir, dbname = check_index(dbfile)
cmd = "gsnap -D {0} -d {1}".format(dbdir, dbname)
cmd += " -B 5 -m 0.1 -i 2 -n 3" # memory, mismatch, indel penalty, nhits
if opts.rnaseq:
cmd += " -N 1"
cmd += " -t {0}".format(opts.cpus)
cmd += " --gmap-mode none --nofails"
if readfile.endswith(".gz"):
cmd += " --gunzip"
try:
offset = "sanger" if guessoffset([readfile]) == 33 else "illumina"
cmd += " --quality-protocol {0}".format(offset)
except AssertionError:
pass
cmd += " " + " ".join(args[1:])
sh(cmd, outfile=gsnapfile, errfile=logfile)
if opts.snp:
snp([gsnapfile, "--cpus={0}".format(opts.cpus)])
return gsnapfile, logfile
def gmap(args):
"""
%prog gmap database.fasta fastafile
Wrapper for `gmap`.
"""
p = OptionParser(gmap.__doc__)
p.add_option("--cross",
default=False,
action="store_true",
help="Cross-species alignment")
p.add_option(
"--npaths",
default=0,
type="int",
help="Maximum number of paths to show."
" If set to 0, prints two paths if chimera"
" detected, else one.",
)
p.set_cpus()
opts, args = p.parse_args(args)
if len(args) != 2:
sys.exit(not p.print_help())
dbfile, fastafile = args
assert op.exists(dbfile) and op.exists(fastafile)
prefix = get_prefix(fastafile, dbfile)
logfile = prefix + ".log"
gmapfile = prefix + ".gmap.gff3"
if not need_update((dbfile, fastafile), gmapfile):
logging.error("`{0}` exists. `gmap` already run.".format(gmapfile))
else:
dbdir, dbname = check_index(dbfile)
cmd = "gmap -D {0} -d {1}".format(dbdir, dbname)
cmd += " -f 2 --intronlength=100000" # Output format 2
cmd += " -t {0}".format(opts.cpus)
cmd += " --npaths {0}".format(opts.npaths)
if opts.cross:
cmd += " --cross-species"
cmd += " " + fastafile
sh(cmd, outfile=gmapfile, errfile=logfile)
return gmapfile, logfile
def gmap(args):
"""
%prog gmap database.fasta fastafile
Wrapper for `gmap`.
"""
p = OptionParser(gmap.__doc__)
p.add_option("--cross", default=False, action="store_true",
help="Cross-species alignment")
p.add_option("--npaths", default=0, type="int",
help="Maximum number of paths to show."
" If set to 0, prints two paths if chimera"
" detected, else one.")
p.set_cpus()
opts, args = p.parse_args(args)
if len(args) != 2:
sys.exit(not p.print_help())
dbfile, fastafile = args
assert op.exists(dbfile) and op.exists(fastafile)
prefix = get_prefix(fastafile, dbfile)
logfile = prefix + ".log"
gmapfile = prefix + ".gmap.gff3"
if not need_update((dbfile, fastafile), gmapfile):
logging.error("`{0}` exists. `gmap` already run.".format(gmapfile))
else:
dbdir, dbname = check_index(dbfile)
cmd = "gmap -D {0} -d {1}".format(dbdir, dbname)
cmd += " -f 2 --intronlength=100000" # Output format 2
cmd += " -t {0}".format(opts.cpus)
cmd += " --npaths {0}".format(opts.npaths)
if opts.cross:
cmd += " --cross-species"
cmd += " " + fastafile
sh(cmd, outfile=gmapfile, errfile=logfile)
return gmapfile, logfile
def snpflow(args):
"""
%prog snpflow trimmed reference.fasta
Run SNP calling pipeline until allele_counts are generated. This includes
generation of native files, SNP_Het file. Speedup for fragmented genomes
are also supported.
"""
p = OptionParser(snpflow.__doc__)
p.set_fastq_names()
p.set_cpus()
opts, args = p.parse_args(args)
if len(args) != 2:
sys.exit(not p.print_help())
trimmed, ref = args
nseqs = len(Fasta(ref))
supercat = nseqs >= 1000
if supercat:
logging.debug("Total seqs in ref: {0} (supercat={1})".\
format(nseqs, supercat))
reads, samples = scan_read_files(trimmed, opts.names)
# Set up directory structure
nativedir, countsdir = "native", "allele_counts"
for d in (nativedir, countsdir):
mkdir(d)
mm = MakeManager()
# Step 0 - index database
db = op.join(*check_index(ref, supercat=supercat, go=False))
cmd = "python -m jcvi.apps.gmap index {0}".format(ref)
if supercat:
cmd += " --supercat"
coordsfile = db + ".coords"
supercatfile = ref.rsplit(".", 1)[0] + ".supercat.fasta"
mm.add(ref, (db, coordsfile), cmd)
else:
mm.add(ref, db, cmd)
# Step 1 - GSNAP alignment and conversion to native file
allnatives = []
allsamstats = []
gmapdb = supercatfile if supercat else ref
for f in reads:
prefix = get_prefix(f, ref)
gsnapfile = op.join(nativedir, prefix + ".gsnap")
nativefile = op.join(nativedir, prefix + ".unique.native")
samstatsfile = op.join(nativedir, prefix + ".unique.sam.stats")
cmd = "python -m jcvi.apps.gmap align {0} {1}".format(gmapdb, f)
cmd += " --outdir={0} --native --cpus=1".format(nativedir)
mm.add((f, db), nativefile, cmd)
cmd = "python -m jcvi.apps.gmap bam {0} {1} --cpus=1".\
format(gsnapfile, gmapdb)
mm.add(nativefile, samstatsfile, cmd)
allnatives.append(nativefile)
allsamstats.append(samstatsfile)
# Step 2 - call SNP discovery
if supercat:
nativeconverted = nativedir + "-converted"
mkdir(nativeconverted)
allnativesc = [op.join(nativeconverted, op.basename(x)) for x in allnatives]
cmd = "tGBS-Convert_Pseudo_Genome_NATIVE_Coordinates.pl"
cmd += " -i {0}/*.native -o {1}".format(nativedir, nativeconverted)
cmd += " -c {0}".format(coordsfile)
cmds = ["rm -rf {0}".format(nativeconverted), cmd]
mm.add(allnatives + [coordsfile], allnativesc, cmds)
runfile = "speedup.sh"
write_file(runfile, speedupsh.format(nativeconverted, opts.cpus))
nativedir = nativeconverted
allsnps = [op.join(nativedir, "{0}.SNPs_Het.txt".format(x)) for x in samples]
mm.add(allnativesc, allsnps, "./{0}".format(runfile))
else:
for s in samples:
snpfile = op.join(nativedir, "{0}.SNPs_Het.txt".format(s))
cmd = "SNP_Discovery-short.pl"
cmd += " -native {0}/{1}.*unique.native".format(nativedir, s)
cmd += " -o {0} -a 2 -ac 0.3 -c 0.8".format(snpfile)
flist = [x for x in allnatives if op.basename(x).split(".")[0] == s]
mm.add(flist, snpfile, cmd)
# Step 3 - generate equal file
allsnps = [op.join(nativedir, "{0}.SNPs_Het.txt".format(x)) for x in samples]
for s in samples:
equalfile = op.join(nativedir, "{0}.equal".format(s))
cmd = "extract_reference_alleles.pl"
cmd += " --native {0}/{1}.*unique.native".format(nativedir, s)
cmd += " --genotype {0}/{1}.SNPs_Het.txt".format(nativedir, s)
cmd += " --allgenotypes {0}/*.SNPs_Het.txt".format(nativedir)
cmd += " --fasta {0} --output {1}".format(ref, equalfile)
mm.add(allsnps, equalfile, cmd)
# Step 4 - generate snp matrix
allequals = [op.join(nativedir, "{0}.equal".format(x)) for x in samples]
matrix = "snps.matrix.txt"
cmd = "generate_matrix.pl"
cmd += " --tables {0}/*SNPs_Het.txt --equal {0}/*equal".format(nativedir)
cmd += " --fasta {0} --output {1}".format(ref, matrix)
mm.add(allsnps + allequals, matrix, cmd)
# Step 5 - generate allele counts
allcounts = []
for s in samples:
allele_counts = op.join(countsdir, "{0}.SNPs_Het.allele_counts".format(s))
cmd = "count_reads_per_allele.pl -m snps.matrix.txt"
cmd += " -s {0} --native {1}/{0}.*unique.native".format(s, nativedir)
cmd += " -o {0}".format(allele_counts)
mm.add(matrix, allele_counts, cmd)
allcounts.append(allele_counts)
# Step 6 - generate raw snps
rawsnps = "Genotyping.H3.txt"
cmd = "/home/shared/scripts/delin/SamplesGenotyping.pl --h**o 3"
cmd += " -pf allele_counts -f {0} --outfile {1}".format(countsdir, rawsnps)
cmds = ["rm -f {0}".format(rawsnps), cmd]
mm.add(allcounts, rawsnps, cmds)
# Step 7 - generate alignment report
sam_summary = "sam.summary"
cmd = "/home/shared/scripts/eddyyeh/alignment_stats.pl"
cmd += " -f {0} -o {1}".format(" ".join(allsamstats), sam_summary)
mm.add(allsamstats, sam_summary, cmd)
native_summary = "native.summary"
cmd = "/home/shared/scripts/eddyyeh/alignment_stats.pl"
cmd += " -n {0} -o {1}".format(" ".join(allnatives), native_summary)
mm.add(allnatives, native_summary, cmd)
mm.write()
def align(args):
"""
%prog align database.fasta read1.fq read2.fq
Wrapper for `gsnap` single-end or paired-end, depending on the number of
args.
"""
from jcvi.formats.fastq import guessoffset
p = OptionParser(align.__doc__)
p.add_option("--rnaseq", default=False, action="store_true",
help="Input is RNA-seq reads, turn splicing on")
p.add_option("--native", default=False, action="store_true",
help="Convert GSNAP output to NATIVE format")
p.set_home("eddyyeh")
p.set_outdir()
p.set_cpus()
opts, args = p.parse_args(args)
if len(args) == 2:
logging.debug("Single-end alignment")
elif len(args) == 3:
logging.debug("Paired-end alignment")
else:
sys.exit(not p.print_help())
dbfile, readfile = args[:2]
outdir = opts.outdir
assert op.exists(dbfile) and op.exists(readfile)
prefix = get_prefix(readfile, dbfile)
logfile = op.join(outdir, prefix + ".log")
gsnapfile = op.join(outdir, prefix + ".gsnap")
nativefile = gsnapfile.rsplit(".", 1)[0] + ".unique.native"
if not need_update((dbfile, readfile), gsnapfile):
logging.error("`{0}` exists. `gsnap` already run.".format(gsnapfile))
else:
dbdir, dbname = check_index(dbfile)
cmd = "gsnap -D {0} -d {1}".format(dbdir, dbname)
cmd += " -B 5 -m 0.1 -i 2 -n 3" # memory, mismatch, indel penalty, nhits
if opts.rnaseq:
cmd += " -N 1"
cmd += " -t {0}".format(opts.cpus)
cmd += " --gmap-mode none --nofails"
if readfile.endswith(".gz"):
cmd += " --gunzip"
try:
offset = "sanger" if guessoffset([readfile]) == 33 else "illumina"
cmd += " --quality-protocol {0}".format(offset)
except AssertionError:
pass
cmd += " " + " ".join(args[1:])
sh(cmd, outfile=gsnapfile, errfile=logfile)
if opts.native:
EYHOME = opts.eddyyeh_home
if need_update(gsnapfile, nativefile):
cmd = op.join(EYHOME, "convert2native.pl")
cmd += " --gsnap {0} -o {1}".format(gsnapfile, nativefile)
cmd += " -proc {0}".format(opts.cpus)
sh(cmd)
return gsnapfile, logfile
def align(args):
"""
%prog align database.fasta read1.fq [read2.fq]
Wrapper for `bowtie2` single-end or paired-end, depending on the number of args.
"""
from jcvi.formats.fastq import guessoffset
p = OptionParser(align.__doc__)
p.set_firstN(firstN=0)
p.add_option("--full", default=False, action="store_true", help="Enforce end-to-end alignment [default: local]")
p.add_option("--reorder", default=False, action="store_true", help="Keep the input read order [default: %default]")
p.set_cutoff(cutoff=800)
p.set_mateorientation(mateorientation="+-")
p.set_sam_options(bowtie=True)
opts, args = p.parse_args(args)
extra = opts.extra
mo = opts.mateorientation
if mo == "+-":
extra += ""
elif mo == "-+":
extra += "--rf"
else:
extra += "--ff"
PE = True
if len(args) == 2:
logging.debug("Single-end alignment")
PE = False
elif len(args) == 3:
logging.debug("Paired-end alignment")
else:
sys.exit(not p.print_help())
firstN = opts.firstN
mapped = opts.mapped
unmapped = opts.unmapped
gl = "--end-to-end" if opts.full else "--local"
dbfile, readfile = args[0:2]
dbfile = get_abs_path(dbfile)
safile = check_index(dbfile)
prefix = get_prefix(readfile, dbfile)
samfile, mapped, unmapped = get_samfile(
readfile, dbfile, bowtie=True, mapped=mapped, unmapped=unmapped, bam=opts.bam
)
logfile = prefix + ".log"
offset = guessoffset([readfile])
if not need_update(safile, samfile):
logging.error("`{0}` exists. `bowtie2` already run.".format(samfile))
return samfile, logfile
cmd = "bowtie2 -x {0}".format(dbfile)
if PE:
r1, r2 = args[1:3]
cmd += " -1 {0} -2 {1}".format(r1, r2)
cmd += " --maxins {0}".format(opts.cutoff)
mtag, utag = "--al-conc", "--un-conc"
else:
cmd += " -U {0}".format(readfile)
mtag, utag = "--al", "--un"
if mapped:
cmd += " {0} {1}".format(mtag, mapped)
if unmapped:
cmd += " {0} {1}".format(utag, unmapped)
if firstN:
cmd += " --upto {0}".format(firstN)
cmd += " -p {0}".format(opts.cpus)
cmd += " --phred{0}".format(offset)
cmd += " {0}".format(gl)
if opts.reorder:
cmd += " --reorder"
cmd += " {0}".format(extra)
# Finally the log
cmd += " 2> {0}".format(logfile)
cmd = output_bam(cmd, samfile)
sh(cmd)
print >>sys.stderr, open(logfile).read()
return samfile, logfile
def align(args):
"""
%prog align database.fasta read1.fq [read2.fq]
Wrapper for `bowtie2` single-end or paired-end, depending on the number of args.
"""
from jcvi.formats.fastq import guessoffset
p = OptionParser(align.__doc__)
p.set_firstN(firstN=0)
p.add_option("--full",
default=False,
action="store_true",
help="Enforce end-to-end alignment [default: local]")
p.add_option("--reorder",
default=False,
action="store_true",
help="Keep the input read order [default: %default]")
p.add_option("--null",
default=False,
action="store_true",
help="Do not write to SAM/BAM output")
p.add_option("--fasta",
default=False,
action="store_true",
help="Query reads are FASTA")
p.set_cutoff(cutoff=800)
p.set_mateorientation(mateorientation="+-")
p.set_sam_options(bowtie=True)
opts, args = p.parse_args(args)
extra = opts.extra
mo = opts.mateorientation
if mo == '+-':
extra += ""
elif mo == '-+':
extra += "--rf"
else:
extra += "--ff"
PE = True
if len(args) == 2:
logging.debug("Single-end alignment")
PE = False
elif len(args) == 3:
logging.debug("Paired-end alignment")
else:
sys.exit(not p.print_help())
firstN = opts.firstN
mapped = opts.mapped
unmapped = opts.unmapped
fasta = opts.fasta
gl = "--end-to-end" if opts.full else "--local"
dbfile, readfile = args[0:2]
dbfile = check_index(dbfile)
prefix = get_prefix(readfile, dbfile)
samfile, mapped, unmapped = get_samfile(readfile,
dbfile,
bowtie=True,
mapped=mapped,
unmapped=unmapped,
bam=opts.bam)
logfile = prefix + ".log"
if not fasta:
offset = guessoffset([readfile])
if not need_update(dbfile, samfile):
logging.error("`{0}` exists. `bowtie2` already run.".format(samfile))
return samfile, logfile
cmd = "bowtie2 -x {0}".format(dbfile)
if PE:
r1, r2 = args[1:3]
cmd += " -1 {0} -2 {1}".format(r1, r2)
cmd += " --maxins {0}".format(opts.cutoff)
mtag, utag = "--al-conc", "--un-conc"
else:
cmd += " -U {0}".format(readfile)
mtag, utag = "--al", "--un"
if mapped:
cmd += " {0} {1}".format(mtag, mapped)
if unmapped:
cmd += " {0} {1}".format(utag, unmapped)
if firstN:
cmd += " --upto {0}".format(firstN)
cmd += " -p {0}".format(opts.cpus)
if fasta:
cmd += " -f"
else:
cmd += " --phred{0}".format(offset)
cmd += " {0}".format(gl)
if opts.reorder:
cmd += " --reorder"
cmd += " {0}".format(extra)
# Finally the log
cmd += " 2> {0}".format(logfile)
if opts.null:
samfile = "/dev/null"
cmd = output_bam(cmd, samfile)
sh(cmd)
print(open(logfile).read(), file=sys.stderr)
return samfile, logfile
def align(args):
"""
%prog align database.fasta read1.fq read2.fq
Wrapper for `gsnap` single-end or paired-end, depending on the number of
args.
"""
from jcvi.formats.fastq import guessoffset
p = OptionParser(align.__doc__)
p.add_option("--rnaseq",
default=False,
action="store_true",
help="Input is RNA-seq reads, turn splicing on")
p.add_option("--native",
default=False,
action="store_true",
help="Convert GSNAP output to NATIVE format")
p.set_home("eddyyeh")
p.set_outdir()
p.set_cpus()
opts, args = p.parse_args(args)
if len(args) == 2:
logging.debug("Single-end alignment")
elif len(args) == 3:
logging.debug("Paired-end alignment")
else:
sys.exit(not p.print_help())
dbfile, readfile = args[:2]
outdir = opts.outdir
assert op.exists(dbfile) and op.exists(readfile)
prefix = get_prefix(readfile, dbfile)
logfile = op.join(outdir, prefix + ".log")
gsnapfile = op.join(outdir, prefix + ".gsnap")
nativefile = gsnapfile.rsplit(".", 1)[0] + ".unique.native"
if not need_update((dbfile, readfile), gsnapfile):
logging.error("`{0}` exists. `gsnap` already run.".format(gsnapfile))
else:
dbdir, dbname = check_index(dbfile)
cmd = "gsnap -D {0} -d {1}".format(dbdir, dbname)
cmd += " -B 5 -m 0.1 -i 2 -n 3" # memory, mismatch, indel penalty, nhits
if opts.rnaseq:
cmd += " -N 1"
cmd += " -t {0}".format(opts.cpus)
cmd += " --gmap-mode none --nofails"
if readfile.endswith(".gz"):
cmd += " --gunzip"
try:
offset = "sanger" if guessoffset([readfile]) == 33 else "illumina"
cmd += " --quality-protocol {0}".format(offset)
except AssertionError:
pass
cmd += " " + " ".join(args[1:])
sh(cmd, outfile=gsnapfile, errfile=logfile)
if opts.native:
EYHOME = opts.eddyyeh_home
if need_update(gsnapfile, nativefile):
cmd = op.join(EYHOME, "convert2native.pl")
cmd += " --gsnap {0} -o {1}".format(gsnapfile, nativefile)
cmd += " -proc {0}".format(opts.cpus)
sh(cmd)
return gsnapfile, logfile
def align(args):
"""
%prog align database.fasta read1.fq read2.fq
Wrapper for `gsnap` single-end or paired-end, depending on the number of
args.
"""
from jcvi.formats.fasta import join
from jcvi.formats.fastq import guessoffset
from jcvi.projects.tgbs import snp
p = OptionParser(align.__doc__)
p.add_option("--join",
default=False,
action="store_true",
help="Join sequences with padded 50Ns")
p.add_option("--rnaseq",
default=False,
action="store_true",
help="Input is RNA-seq reads, turn splicing on")
p.add_option("--snp",
default=False,
action="store_true",
help="Call SNPs after GSNAP")
p.set_cpus()
opts, args = p.parse_args(args)
if len(args) == 2:
logging.debug("Single-end alignment")
elif len(args) == 3:
logging.debug("Paired-end alignment")
else:
sys.exit(not p.print_help())
dbfile, readfile = args[0:2]
if opts.join:
dbfile = join([dbfile, "--gapsize=50", "--newid=chr1"])
assert op.exists(dbfile) and op.exists(readfile)
prefix = get_prefix(readfile, dbfile)
logfile = prefix + ".log"
gsnapfile = prefix + ".gsnap"
if not need_update((dbfile, readfile), gsnapfile):
logging.error("`{0}` exists. `gsnap` already run.".format(gsnapfile))
else:
dbdir, dbname = check_index(dbfile)
cmd = "gsnap -D {0} -d {1}".format(dbdir, dbname)
cmd += " -B 5 -m 0.1 -i 2 -n 3" # memory, mismatch, indel penalty, nhits
if opts.rnaseq:
cmd += " -N 1"
cmd += " -t {0}".format(opts.cpus)
cmd += " --gmap-mode none --nofails"
if readfile.endswith(".gz"):
cmd += " --gunzip"
try:
offset = "sanger" if guessoffset([readfile]) == 33 else "illumina"
cmd += " --quality-protocol {0}".format(offset)
except AssertionError:
pass
cmd += " " + " ".join(args[1:])
sh(cmd, outfile=gsnapfile, errfile=logfile)
if opts.snp:
snp([gsnapfile, "--cpus={0}".format(opts.cpus)])
return gsnapfile, logfile
def snpflow(args):
"""
%prog snpflow trimmed reference.fasta
Run SNP calling pipeline until allele_counts are generated. This includes
generation of native files, SNP_Het file. Speedup for fragmented genomes
are also supported.
"""
p = OptionParser(snpflow.__doc__)
p.set_fastq_names()
p.set_cpus()
opts, args = p.parse_args(args)
if len(args) != 2:
sys.exit(not p.print_help())
trimmed, ref = args
nseqs = len(Fasta(ref))
supercat = nseqs >= 1000
if supercat:
logging.debug("Total seqs in ref: {0} (supercat={1})".\
format(nseqs, supercat))
reads, samples = scan_read_files(trimmed, opts.names)
# Set up directory structure
nativedir, countsdir = "native", "allele_counts"
for d in (nativedir, countsdir):
mkdir(d)
mm = MakeManager()
# Step 0 - index database
db = op.join(*check_index(ref, supercat=supercat, go=False))
cmd = "python -m jcvi.apps.gmap index {0}".format(ref)
if supercat:
cmd += " --supercat"
coordsfile = db + ".coords"
supercatfile = ref.rsplit(".", 1)[0] + ".supercat.fasta"
mm.add(ref, (db, coordsfile), cmd)
else:
mm.add(ref, db, cmd)
# Step 1 - GSNAP alignment and conversion to native file
allnatives = []
allsamstats = []
gmapdb = supercatfile if supercat else ref
for f in reads:
prefix = get_prefix(f, ref)
gsnapfile = op.join(nativedir, prefix + ".gsnap")
nativefile = op.join(nativedir, prefix + ".unique.native")
samstatsfile = op.join(nativedir, prefix + ".unique.sam.stats")
cmd = "python -m jcvi.apps.gmap align {0} {1}".format(gmapdb, f)
cmd += " --outdir={0} --native --cpus=1".format(nativedir)
mm.add((f, db), nativefile, cmd)
cmd = "python -m jcvi.apps.gmap bam {0} {1} --cpus=1".\
format(gsnapfile, gmapdb)
mm.add(nativefile, samstatsfile, cmd)
allnatives.append(nativefile)
allsamstats.append(samstatsfile)
# Step 2 - call SNP discovery
if supercat:
nativeconverted = nativedir + "-converted"
mkdir(nativeconverted)
allnativesc = [op.join(nativeconverted, op.basename(x)) for x in allnatives]
cmd = "tGBS-Convert_Pseudo_Genome_NATIVE_Coordinates.pl"
cmd += " -i {0}/*.native -o {1}".format(nativedir, nativeconverted)
cmd += " -c {0}".format(coordsfile)
cmds = ["rm -rf {0}".format(nativeconverted), cmd]
mm.add(allnatives + [coordsfile], allnativesc, cmds)
runfile = "speedup.sh"
write_file(runfile, speedupsh.format(nativeconverted, opts.cpus))
nativedir = nativeconverted
allsnps = [op.join(nativedir, "{0}.SNPs_Het.txt".format(x)) for x in samples]
mm.add(allnativesc, allsnps, "./{0}".format(runfile))
else:
for s in samples:
snpfile = op.join(nativedir, "{0}.SNPs_Het.txt".format(s))
cmd = "SNP_Discovery-short.pl"
cmd += " -native {0}/{1}.*unique.native".format(nativedir, s)
cmd += " -o {0} -a 2 -ac 0.3 -c 0.8".format(snpfile)
flist = [x for x in allnatives if op.basename(x).split(".")[0] == s]
mm.add(flist, snpfile, cmd)
# Step 3 - generate equal file
allsnps = [op.join(nativedir, "{0}.SNPs_Het.txt".format(x)) for x in samples]
for s in samples:
equalfile = op.join(nativedir, "{0}.equal".format(s))
cmd = "extract_reference_alleles.pl"
cmd += " --native {0}/{1}.*unique.native".format(nativedir, s)
cmd += " --genotype {0}/{1}.SNPs_Het.txt".format(nativedir, s)
cmd += " --allgenotypes {0}/*.SNPs_Het.txt".format(nativedir)
cmd += " --fasta {0} --output {1}".format(ref, equalfile)
mm.add(allsnps, equalfile, cmd)
# Step 4 - generate snp matrix
allequals = [op.join(nativedir, "{0}.equal".format(x)) for x in samples]
matrix = "snps.matrix.txt"
cmd = "generate_matrix.pl"
cmd += " --tables {0}/*SNPs_Het.txt --equal {0}/*equal".format(nativedir)
cmd += " --fasta {0} --output {1}".format(ref, matrix)
mm.add(allsnps + allequals, matrix, cmd)
# Step 5 - generate allele counts
allcounts = []
for s in samples:
allele_counts = op.join(countsdir, "{0}.SNPs_Het.allele_counts".format(s))
cmd = "count_reads_per_allele.pl -m snps.matrix.txt"
cmd += " -s {0} --native {1}/{0}.*unique.native".format(s, nativedir)
cmd += " -o {0}".format(allele_counts)
mm.add(matrix, allele_counts, cmd)
allcounts.append(allele_counts)
# Step 6 - generate raw snps
rawsnps = "Genotyping.H3.txt"
cmd = "/home/shared/scripts/delin/SamplesGenotyping.pl --h**o 3"
cmd += " -pf allele_counts -f {0} --outfile {1}".format(countsdir, rawsnps)
cmds = ["rm -f {0}".format(rawsnps), cmd]
mm.add(allcounts, rawsnps, cmds)
# Step 7 - generate alignment report
sam_summary = "sam.summary"
cmd = "/home/shared/scripts/eddyyeh/alignment_stats.pl"
cmd += " -f {0} -o {1}".format(" ".join(allsamstats), sam_summary)
mm.add(allsamstats, sam_summary, cmd)
native_summary = "native.summary"
cmd = "/home/shared/scripts/eddyyeh/alignment_stats.pl"
cmd += " -n {0} -o {1}".format(" ".join(allnatives), native_summary)
mm.add(allnatives, native_summary, cmd)
mm.write()
