Paired Samples in Lima Voom for Bulk RNAseq


I am not sure if I am suppose to ask this question here or not. Seems like most topics are about server/tool errors. If I am in the wrong place, please let me know where I need to go.

I have been trying to analyze Bulk-RNAseq data following the Reads to Count tutorial by Maria Doyle et al. Everything went well, but the results are a bit funky. I believe it is because my data comes from 3 different cell types. The conditons are as follows: Cell1, Cell2, Cell3 and Cell1 treated, Cell2 treated, and Cell3 treated. I am only interested in the “treated” effect, rather than the variation within each cell type. Is there a way to “pair” the differential expression to between the cell1 and cell1 treated, but still compare the untreated to treated as a groups?