I am working with an RNA-seq dataset that combines samples from strand-specific and unstranded technology. Ideally I would like to use all the samples to be able to detect DE genes between biological groups.
I am trying to deal with the problem at the gene expression level and treating the difference in technologies as batch effect, however my dataset is no very balanced (the biologic groups are not evenly distributed between the two technologies).
Has anyone dealt with a similar situation before? Is there a way to account for the difference in technologies in upstream stages (like counting)?
Thanks a lot,