def work_script(self):
return '''#!/bin/bash
source jre-8u92
source gatk-3.6.0
gatk='{gatk}'
set -eo pipefail
$gatk -T -T SelectVariants -V {input} -R {reference} --selectTypeToInclude MNP \
--selectTypeToInclude MIXED \
--out {output}.temp.vcf.gz
mv {output}.temp.vcf.gz {output}
'''.format(input=self.input().path,
output=self.output().path,
reference=self.reference,
gatk=gatk.format(mem=self.mem * self.n_cpu))
def work_script(self):
return '''#!/bin/bash
source jre-8u92
source gatk-3.6.0
gatk='{gatk}'
set -eo pipefail
$gatk -T GenotypeGVCFs -R {reference} -L {intervals} -o {output}.temp.vcf.gz --includeNonVariantSites {variants}
mv {output}.temp.vcf.gz {output}
'''.format(output=self.output().path,
intervals=self.input()[0].path,
gatk=gatk.format(mem=self.mem * self.n_cpu),
reference=self.reference,
variants="\\\n".join([
" --variant " + lib.path
for lib in self.input()[1]
]))
def work_script(self):
return '''#!/bin/bash
source jre-8u92
gatk='{gatk}'
set -euo pipefail
$gatk -T HaplotypeCaller \
-R {reference} \
-I {input} \
-dontUseSoftClippedBases\
--emitRefConfidence GVCF \
-o {output}.temp.g.vcf
mv {output}.temp.g.vcf {output}
'''.format(input=self.input().path,
output=self.output().path,
gatk=gatk.format(mem=self.mem * self.n_cpu),
reference=self.reference)
def work_script(self):
return '''#!/bin/bash
source jre-8u92
source gatk-3.6.0
gatk='{gatk}'
picard='{picard}'
set -euo pipefail
$picard BuildBamIndex VERBOSITY=ERROR QUIET=true I={input}
$gatk -T SplitNCigarReads --logging_level ERROR -R {reference} -I {input} -o {output}.temp -rf ReassignOneMappingQuality -RMQF 255 -RMQT 60 -U ALLOW_N_CIGAR_READS
mv {output}.temp.bai {output}.bai
mv {output}.temp {output}
'''.format(input=self.input().path,
output=self.output().path,
picard=picard.format(mem=self.mem * self.n_cpu),
gatk=gatk.format(mem=self.mem * self.n_cpu),
reference=self.reference)
def work_script(self):
recal = os.path.join(self.base_dir, VERSION, PIPELINE, self.library,
self.library + "_recal.tsv")
return '''#!/bin/bash
source jre-8u92
source gatk-3.6.0
gatk='{gatk}'
set -euo pipefail
$gatk -T BaseRecalibrator -R {reference} -I {input} -knownSites {snp_db} -o {recal}
$gatk -T PrintReads -R {reference} -I {input} -BQSR {recal} -o {output}.temp
mv {output}.temp {output}
'''.format(gatk=gatk.format(mem=self.mem * self.n_cpu),
input=self.input().path,
output=self.output().path,
reference=self.reference,
recal=recal)
def work_script(self):
return '''#!/bin/bash
source jre-8u92
source gatk-3.6.0
source vcftools-0.1.13;
gatk='{gatk}'
set -eo pipefail
$gatk -T SelectVariants -V {input} -R {reference} \
--selectTypeToInclude NO_VARIATION \
--selectTypeToInclude SNP \
--out {output}.temp.vcf.gz
# Filter out * which represents spanning deletions
gzip -cd {output}.temp.vcf.gz | grep -v $'[,\t]\*' | bgzip -c > {output}.temp2.vcf.gz
mv {output}.temp2.vcf.gz {output}
'''.format(input=self.input().path,
output=self.output().path,
reference=self.reference,
gatk=gatk.format(mem=self.mem * self.n_cpu))
def work_script(self):
perfile = math.ceil(len(self.input()) / self.N_gvcfs)
start_idx = perfile * self.idx
end_idx = perfile * (self.idx + 1)
self.variants = self.input(
)[start_idx:end_idx] if self.idx < self.N_gvcfs - 1 else self.input(
)[start_idx:]
return '''#!/bin/bash
source jre-8u92
source gatk-3.6.0
gatk='{gatk}'
set -eo pipefail
$gatk -T CombineGVCFs -R {reference} -o {output}.temp {variants}
mv {output}.temp {output}
'''.format(output=self.output().path,
gatk=gatk.format(mem=self.mem * self.n_cpu),
reference=self.reference,
variants="\\\n".join([
" --variant " + lib.path
for lib in self.variants
]))