I am using bulk RNA-seq data (rRNA depleted) from cell lines infected with RNA virus. RNA was extracted at different time intervals post viral infection. After QC and STAR/HISAT2 alignment against hg38 genome, I observed poor reads mapping percentage with both the aligners, specifically in samples collected after 24hr post viral infection. In viral infected samples majority of the RNA reads mapped with viral genome (20% human and almost 70% viral genome) as compared to the uninfected samples with >90% hg38 mapping.
I need to compare human/host gene expression profiles of uninfected cells (90% human) with infected cells (20% human + 70% viral reads). What should be the best way to normalize this huge difference in the number of host specific reads in both the samples?
This is relatively a new field for me, I would be glad if someone could help me figure out a solution.
