The frontline support team will be unavailable to answer questions on April 15th and 17th 2019. We will be back soon after. Thank you for your patience and we apologize for any inconvenience!
Calling variants in non model organisms
Dear Gatk team,
I've been going through your best practices workflow to clean up my DNAseq data. I have 4 whole genomes (20x depth) from two grape cultivars, two genomes from each cultivar, and we would like to find variants that allow us to distinguish among these two cultivars. I would like to know if it make sense, given my dataset, to generate gVCF files for each genome and then do the jointGenotyping for each cultivar. Or would you recommend to run a plane HaplotypeCaller for each of the four available genomes, without -ERC GVCF, and afterwards perform a VQSR on each of the .vcf files. Is there any reason to think that the latter option would retain more variants (i.e. singletons).
Thank you very much in advanced.