Read group addition order in RNA-seq short variant discovery (SNPs + Indels)

Question

[iamsmor]

Hello, I’m following the GATK best practices for RNA-seq short variant discovery (SNPs + Indels) and wondering about the correct point to add Read Groups (RGs). Does the order (before/after MarkDuplicates or SplitNCigarReads) matter for RNA-seq variant calling with GATK (HaplotypeCaller)? Any official clarification or reference from the GATK team or papers? Pipeline: HISAT2, AddOrReplaceReadGroups, MarkDuplicates, SplitNCigarReads, BaseRecalibrator, HaplotypeCaller Thank you for any kind of help

Answer 1

[GenoMax]

Ideally you would do this at the beginning of the process as you align the data. Read groups would be tagging the samples with relevant info e.g. what lane they were in, if they ran on multiple flowcells etc (all of this may not be applicable in your case).

See Expected Input section: https://gatk.broadinstitute.org/hc/en-us/articles/360035535912-Data-pre-processing-for-variant-discovery

[iamsmor]

ı didnot do when ı align rnaseq datasets but I did it as soon as I realized I had to do it. ı am following https://gatk.broadinstitute.org/hc/en-us/articles/360035531192-RNAseq-short-variant-discovery-SNPs-Indels but here is not exact information.

[GenoMax]

Do it before you start doing any manipulations after alignments.

[iamsmor]

yes this is exactly what ı have done but anway ı couldnot find any clear reference for exact pipeline .so actually ı write here maybe someone can help for reference , because nothing clear when ı search on google. thank you for help, ı also feel ı am right but ı amnot sure anyway without papers.

[GenoMax]

https://gatk.broadinstitute.org/hc/en-us/articles/360035531192-RNAseq-short-variant-discovery-SNPs-Indels references the https://gatk.broadinstitute.org/hc/en-us/articles/360035535912-Data-pre-processing-for-variant-discovery pipeline linked above. In RNAseq pipeline STAR is replacing default aligner (to account for splicing) so the rest of the steps from there on should be as per the input requirements for DNA pipeline.

We use MergeBamAlignment and MarkDuplicates (similarly to our DNA pre-processing best practices pipeline)

[iamsmor]

thank you very much. actually ı follow this link's pipeline also but should read more carefully for meaning . thanks alot Genomax .