def run_gatk():
# generate interval list file
intvfile = {{job.outdir | path.join: "interval.list" | quote}}
cmd = '{samtools} idxstats {tumor!r} | head -1 | cut -f1 > {intvfile!r}'.format(
samtools = samtools,
tumor = tumor,
intvfile = intvfile
)
runcmd(cmd)
mem = mem2(mem, 'java')
params['I:tumor'] = tumor
params['I:normal'] = normal
params.R = ref
params.o = outfile
params.nct = nthread
params.L = intvfile
cmd = '{gatk} -T MuTect2 {mem} -Djava.io.tmpdir={tmpdir!r} {args}'.format(
gatk = gatk,
mem = mem,
tmpdir = tmpdir,
args = cmdargs(params, dash = '-', equal = ' ')
)
runcmd(cmd)
if gz: runcmd(['gzip', outfile])
def run_picard():
global infile
mem = mem2(argsmem, '-jdict')
mem['-Djava.io.tmpdir'] = tmpdir
shellpicard = Shell(subcmd = True, dash = '', equal = '=').picard(**mem)
if not (steps.sort or steps.index or steps.markdup or steps.rmdup):
shellpicard.SamFormatConverter(TMP_DIR = tmpdir, I = infile, O = outfile).run()
else:
bamfile = outfile
if steps.sort:
bamfile = path.join(joboutdir, inprefix + '.sorted.bam')
shellpicard.ShortSam(TMP_DIR = tmpdir, I = infile, O = bamfile, SO = sortby).run()
if infile != {{i.infile | quote}}:
shell.rm(f = True, _ = infile)
infile = bamfile
if steps.markdup:
mfile = "/dev/null"
bamfile = path.join(joboutdir, inprefix + '.dedup.bam')
shellpicard.MarkDuplicates(REMOVE_DUPLICATES = 'true' if steps.rmdup else 'false', TMP_DIR = tmpdir, I = infile, O = bamfile, M = mfile).run()
if infile != {{i.infile | quote}}:
shell.rm(f = True, _ = infile)
infile = bamfile
if steps.index:
shellpicard.BuildBamIndex(TMP_DIR = tmpdir, I = infile, O = outfile + '.bai').run()
if infile != outfile:
if path.exists(infile + '.bai'):
shell.mv(infile + '.bai', outfile + '.bai')
shell.mv(infile, outfile)
def run_samtools():
global infile
if not (steps.sort or steps.index or steps.markdup or steps.rmdup):
subshell.samtools.view(b = True, o = outfile, O = 'bam', _ = infile).run()
else:
bamfile = outfile
if steps.sort:
mem = mem2(argsmem, 'M')
bamfile = path.join(joboutdir, inprefix + '.sorted.bam')
subshell.samtools.sort(
m = mem + 'M',
n = sortby == 'queryname',
o = bamfile,
T = tmpdir,
O = 'bam',
_ = infile,
**{'@': nthread}
).run()
if infile != {{i.infile | quote}}:
shell.rm(infile, f = True)
infile = bamfile
if steps.markdup or steps.rmdup:
bamfile = path.join(joboutdir, inprefix + '.dedup.bam')
subshell.rmdup(infile, bamfile).run()
if infile != {{i.infile | quote}}:
shell.rm(infile, f = True)
infile = bamfile
if steps.index:
subshell.samtools.index(bamfile, outfile + '.bai')
if infile != outfile:
if path.exists(infile + '.bai'):
shell.mv(infile + '.bai', outfile + '.bai')
shell.mv(infile, outfile)
def run_virmid():
params.R = ref
params.D = tumor
params.N = normal
params.w = joboutdir
cmd = '{virmid} {mem} -Djava.io.tmpdir={tmpdir!r} {args}'.format(
virmid = virmid,
mem = mem2(mem, 'java'),
tmpdir = tmpdir,
args = cmdargs(params)
)
runcmd(['mv', path.join(joboutdir, '*.virmid.som.passed.vcf'), outfile])
if gz: runcmd(['gzip', outfile])
def run_gatk():
mem = mem2(argsmem, '-jdict')
intfile = path.join(joboutdir, outprefix + '.intervals')
mem['-Djava.io.tmpdir={}'.format(shell.shquote(tmpdir))] = True
gatksh = Shell(equal=' ', dash='-').gatk
rtcparams = params.get('RealignerTargetCreator', Box())
rtcparams.T = 'RealignerTargetCreator'
rtcparams.R = ref
rtcparams.I = infile
rtcparams.o = intfile
rtcparams.nt = nthread
rtcparams._ = list(mem.keys())
gatksh(**rtcparams).run()
bamfileir = path.join(joboutdir, outprefix + '.ir.bam')
irparams = params.get('IndelRealigner', Box())
irparams.T = 'IndelRealigner'
irparams.R = ref
irparams.I = infile
irparams.o = bamfileir
irparams._ = list(mem.keys())
irparams.targetIntervals = intfile
gatksh(**irparams).run()
recaltable = path.join(joboutdir, outprefix + '.recaltable')
brparams = params.get('BaseRecalibrator', Box())
brparams.T = 'BaseRecalibrator'
brparams.R = ref
brparams.I = bamfileir
brparams.o = recaltable
brparams.nct = nthread
brparams._ = list(mem.keys())
brparams.knownSites = knownSites
gatksh(**brparams).run()
prparams = params.get('PrintReads', Box())
prparams.T = 'PrintReads'
prparams.R = ref
prparams.I = bamfileir
prparams.o = outfile
prparams.nct = nthread
prparams._ = list(mem.keys())
gatksh(**prparams).run()
shell.rm(bamfileir, f=True)
shell.mv(outprefix + '.bai', outfile + '.bai')
def run_gatk():
gatkmem = mem2(mem, 'jdict')
gatkmem['Djava.io.tmpdir={!r}'.format(tmpdir)] = True
gatksh = Shell(equal = ' ', dash = '-')
params.T = 'HaplotypeCaller'
params.R = ref
params.I = infile
params.o = outfile
params.nct = nthread
params._ = list(gatkmem.keys())
gatksh(**params).run()
if gz: shell.gzip(outfile)
def run_strelka():
cparams = {{args.configParams | repr}}
cparams.normalBam = normal
cparams.tumorBam = tumor
cparams.referenceFasta = ref
cparams.runDir = joboutdir
runcmd('{strelka} {args}'.format(strelka = strelka, args = cmdargs(cparams)))
params.m = 'local'
params.g = mem2(mem, 'G')[:-1]
params.j = nthread
runcmd('{joboutdir}/runWorkflow.py {args}'.format(joboutdir = joboutdir, args = cmdargs(params)))
snvvcf = path.join(joboutdir, 'results', 'variants', 'somatic.snvs.vcf.gz')
indvcf = path.join(joboutdir, 'results', 'variants', 'somatic.indels.vcf.gz')
_mergeAndAddGT(snvvcf, indvcf, outfile)
if gz: runcmd(['gzip', outfile])
def run_strelka():
# config
cfgParams.bam = infile
cfgParams.referenceFasta = ref
cfgParams.runDir = joboutdir
Shell().strelka(**cfgParams).run()
# run the pipeline
params.m = 'local'
params.j = nthread
params.g = mem2(mem, 'G')[:-1]
Shell({'runWorkflow': path.join(joboutdir, 'runWorkflow.py')}).runWorkflow(**params).run()
# mv output file to desired outfile
ofile = path.join(joboutdir, 'results', 'variants', 'genome.S1.vcf.gz')
shell.mv(ofile, outfile + '.gz')
if not gz: shell.gunzip(outfile + '.gz')
def run_biobambam(): mem = mem2(argsmem, 'M') if steps.index: params.index = 1 params.indexfilename = outfile + '.bai' params.I = infile params.O = outfile params.SO = sortby params.blockme = mem params.tmpfile = path.join(tmpdir, 'biobambam.tmp') params.inputformat = infmt params.outfmt = 'bam' params.inputthreads = nthread params.outputthreads = nthread params.markduplicates = int(steps.markdup) params.rmdup = int(steps.rmdup) Shell(dash = '', equal = '=').biobambam(**params).run()
from sys import stderr
from shutil import move
from pyppl import Box
from bioprocs.utils import runcmd, mem2, cmdargs
params = {{args.params}}
try:
{% case args.tool %}
{% when 'trimmomatic' %}
mem = mem2 ({{args.mem | quote}}, "java")
minlen = str({{args.minlen}} * 2)
adfile = "{{job.outdir}}/adapters.fa"
with open (adfile, "w") as ad:
ad.write (">TruSeq3_IndexedAdapter\n")
ad.write ({{args.adapter | quote}} + "\n")
params['threads'] = {{args.nthread}}
cmd = '{{args.trimmomatic}} %s SE %s "{{in.fq}}" "{{out.outfq}}" ILLUMINACLIP:%s:2:30:10 LEADING:{{args.cut5}} TRAILING:{{args.cut3}} SLIDINGWINDOW:4:{{args.minq}} MINLEN:%s' % (mem, cmdargs(params, dash = '-', equal = ' '), adfile, minlen)
runcmd (cmd)
{% when 'cutadapt' %}
params['a'] = {{args.adapter | quote}}
params['u'] = "{{args.cut5}}"
params['u'] = "-{{args.cut3}}"
params['m'] = {{args.minlen}}
params['q'] = "{{args.minq}},{{args.minq}}"
params['o'] = {{out.outfq | quote}}
cmd = '{{args.cutadapt}} %s "{{in.fq}}"' % cmdargs(params, dash = '-', equal = ' ')
runcmd (cmd)
cmd = '{{args.biobambam}} %s' % cmdargs(params, dash = '', equal = '=')
runcmd (cmd)
{% when 'bedtools' %}
params['i'] = infile
params['fq'] = fqfile1
params['fq2'] = fqfile2
cmd = '{{args.bedtools}} bamtofastq %s' % cmdargs(params, dash = '-', equal = ' ')
runcmd (cmd)
{% when 'samtools' %}
params['t'] = True
params['1'] = fqfile1
params['2'] = fqfile2
cmd = '{{args.samtools}} fastq %s "%s"' % (cmdargs(params, dash = '-', equal = ' '), infile)
runcmd (cmd)
{% when 'picard' %}
mem = mem2({{ args.mem | quote }}, 'Java')
params[mem] = True
params['-Djava.io.tmpdir'] = tmpdir
params['TMP_DIR'] = tmpdir
params['I'] = infile
params['F'] = fqfile1
params['F2'] = fqfile2
cmd = '{{args.picard}} SamToFastq %s' % cmdargs(params, dash='', equal='=')
runcmd (cmd)
{% endcase %}
{% if args.gz %}
runcmd ('gzip "%s"' % (fqfile1))
runcmd ('gzip "%s"' % (fqfile2))
{% endif %}
except Exception as ex:
if not rg['ID']:
g = re.search (r'[^a-zA-Z0-9]+(L\\d+)[^a-zA-Z0-9]+', "{{o.outfile | fn}}")
rg['ID'] = g.group(1) if g else "{{o.outfile | fn}}.L{{job.index}}"
if not rg['SM']:
rg['SM'] = "{{o.outfile | fn}}"
tmpdir = path.join ("{{args.tmpdir}}", "{{proc.id}}.{{i.infile | fn}}.{{job.index}}")
if not path.exists (tmpdir):
makedirs (tmpdir)
params = {{args.params}}
try:
{% case args.tool %}
############## picard
{% when 'picard' %}
mem = mem2({{ args.mem | quote }})
params['-Djava.io.tmpdir'] = tmpdir
params['TMP_DIR'] = tmpdir
params['I'] = {{i.infile | quote}}
params['O'] = {{o.outfile | quote}}
for k,v in rg.items():
params['RG' + k] = v
runcmd ('{{args.picard}} AddOrReplaceReadGroups %s %s' % (mem, cmdargs(params, dash='', equal='=')))
############## bamutil
{% when 'bamutil' %}
params['RG'] = "@RG\\tID:%s\\t%s" % (rg['ID'], "\\t".join([k + ":" + v for k,v in rg.items() if k!='ID']))
params['in'] = {{i.infile | quote}}
params['out'] = {{o.outfile | quote}}
tool = {{args.tool | repr}}
picard = {{args.picard | repr}}
chain = {{args.lochain | repr}}
ref = {{args.ref | repr}}
params = {{args.params | repr}}
mem = {{args.mem | repr}}
tmpdir = {{args.tmpdir | repr}}
if not chain:
logger.error('Chain file (args.lochain) not provided!')
exit(1)
# picard LiftoverVcf -Xmx4g -Xms1g I=TCGA-05-4382-10.vcf O=1.vcf CHAIN=liftovers/hg38ToHg19.over.chain.gz R=ucsc_hg19.fa REJECT=r.vcf
if tool == 'picard':
params.I = infile
params.O = outfile
params.CHAIN = chain
params.REJECT = umfile
params.R = ref
javamem = mem2(mem, 'java')
for jm in javamem.split():
params['-' + jm[1:]] = True
params['-Djava.io.tmpdir'] = tmpdir
cmd = '{picard} LiftoverVcf {params}'
runcmd(
cmd.format(picard=picard, params=cmdargs(params, equal='=', dash='')))