gatk_genotype.sge

#!/bin/bash

#gatk_genotype.sge
#sge submission script to call joint genotypes across samples
#M. Supple
#last modified 9 August 2016

#usage
#qsub gatk_genotype.sge </path/to/gvcf/dir/> <path/to/reference/ref.fa>
	#</path/to/bams/dir/> is a directory with the input g.vcf files
	#<path/to/reference/ref.fa> is the reference sequence
#note: see hardcoded parameters below

#requires
#gatk

#output
#single genotyped VCF


#$ -N gatk_genotype
#$ -o gatk_genotype.output
#$ -q hugemem.q
#$ -l virtual_free=220g,h_vmem=221g
#$ -j y
#$ -cwd

#print some useful sge information
echo Job $JOB_NAME started `date` in queue $QUEUE with id=$JOB_ID on `hostname` 

#get input information
gvcfdir=$1
reference=$2

het=0.005
indelhet=.0005

#determine samples to use
variants=""
samples=($(ls $gvcfdir/S*.g.vcf))
for ((a=0; a<${#samples[@]}; a++))
        do
		variants+="--variant ${samples[a]} "
	done
	echo $variants

#call variants
java -Xmx150G -jar /home/msupple/programs/GenomeAnalysisTK.jar \
	-T GenotypeGVCFs \
	-hets $het -indelHeterozygosity $indelhet \
	-R $reference \
	$variants \
	-o combined.vcf


#print note that job completed
echo done `date`