We've moved!
This site is now read-only. You can find our new documentation site and support forum for posting questions here.
Be sure to read our welcome blog!

HaplotypeCaller - Optimising speed and accuracy

I am using haplotypecaller module GATK I need to generate gVCF files for more than 100 samples. I have gone through already asked questions about increasing the speed of haplotyecaller. What I found was haplotypcaller spark is not a recommended option , increasing the number of cores and threads will effect its accuracy . Importantly, what one can do passing interval argument to call variants in recommended regions (excluding centromeres and telomeres). Depending on these conclusions, I am using this command -

gatk HaplotypeCaller --java-options "-Xmx8G -XX:+UseParallelGC -XX:ParallelGCThreads=4" -R ../reference/GRCh37.fa -I ../BAM_files/test.bam -O test.raw.snps.indels.g.vcf -L b37_wgs_calling_regions.v1.list -ERC GVCF > log_test.txt

Please suggest whether it is fine or I am missing some parameter that can enhance the speed. Thank You


Sign In or Register to comment.