How/when to merge biological replicates during GATK work flow

Question

[Genetics]

When we have to merge biological replicates during GATK work flow if I am trying to identify RNA editing (A to I) sites. Can I use CatVariants? For alignment I used STAR and then followed the GATK. I was trying to find answers to this question in this forum but it is not clear to me.

Any help is appreciated.

[Ram]

Did you already get to VCF? Merging FASTQs or BAMs would be the ideal thing to do I think. At the VCF level, you'll still retain an entry for each replicate even if you merge variant loci.

[Genetics]

I generated a VCF file for one replicate at varient calling step.I have two more replicates. Previously I was thinking to merge at VCF level . But as you have suggested I will try to merge all BAMs and do all the steps below after merging all three bam files. 1) handling splicing events in rna-seq data 2) calling variants by “HaplotypeCaller” function 3) variant selection by “SelectVariants” function, 4) variant filtration by “VariantFiltration” function 5) variant annotation.

Is that correct?please let me know. Thank you.

[Ram]

I mean, it makes sense to me to merge reads and then proceed with the pipeline. I can't really say if it's "correct". Try and learn as you go.