Include plasmid sequences in RNA-seq differential expression analysis for bacteria?

Question

[MaxMin]

Hi everyone,

I’m performing differential expression analysis on bacterial RNA-seq data. My organism has one main chromosome and 2 plasmids.

Should I include the plasmid sequences in the analysis? My current plan is to:

1. Build a combined reference genome (chromosome + plasmids) for read mapping,

1. Run featureCounts using a merged GFF3 annotation file that includes all replicons, and

1. Perform the differential expression analysis (e.g. with DESeq2) on the combined count matrix.

Does this workflow make sense? Or would it be better to analyze plasmid-encoded genes separately from chromosomal ones?

Thanks for any suggestions or best-practice advice!

[GenoMax]

Should I include the plasmid sequences in the analysis?

If you are interested in expression of genes from these plasmids then yes.

Answer 1

[swbarnes2]

Do everything together; align, count, analyze. How would you normalize a handful of plasmid genes anyway?