Hello,

I am using DESeq2 to analyze ATAC-seq data. The samples are from human patients, pre and post therapy -- so, require a paired analysis which can be done by including the donor ID in the model matrix. The samples had to be processed in 3 batches (no known variables are confounded with batch). 

My question: What is the best way to incorporate the pairwise analysis and account for batch effects? Include both in the model matrix? Does the order matter? I don't think I can remove the batch effects with Limma or Combat prior to DESeq2 because it requires raw counts as input.

Any advice would be greatly appreciated! Thank you!

Josephine