Does GATK4 provide better quality of SNP/Indels calls over GATK3.8?
As I understand from the posts in GATK website, the improvement is mainly in spark implementation which enables core parallelization. But other than this, it does not appear to be different in terms of the methodological aspects of variant calling. Anyone have compared the variants called with these two different versions? Noticed any difference? Or the final callset is very similar in the end? The key aspect for me is - should I update my workflow to version 4 from 3.8 if I do not worry about core parallelization? Thanks in advance.
• 718 views
•
link
0 answers
No answers yet.
Log in to answer this question.