I have two RNA-Seq datasets. Both have 2 genotypes and 2 treatment conditions. They both only differ in one of the treatment conditions.
Dataset 1 - Genotype I and Genotype II. Treatment A and treatment B
Dataset 2 - Genotype I and Genotype II. Treatment A and treatment C
The cell lines used are the same, but the sequencing technique is different between the two.
I am assuming that the differential genes and Gene ontology for the comparison - Genotype I treatment A vs Genotype II treatment A from dataset 1 and from dataset 2 would match. I did a PCA first, but the samples for the same genotypes did not coincide -
Ideally, all points from group IA and IIA should be coinciding. Am I doing something wrong? Can we make a comparison like this or not? And if I have to compare these two datasets what do I look at? On what basis can I make a comparison?
I have accounted for the Batch effect. I used DESeq2 for the differential gene expression analysis. The model I use is - Batch + genotype + treatment + genotype:treatment.
Is there a reason we would prefer edgeR over DESeq2?
Sorry, I thought you didn't include the batch variable in the model :-)