Hi,

I have an RNA-seq experiment with the design:

name celltype treatment group WT_Mock_1 WT Mock A WT_Mock_2 WT Mock B WT_Mock_3 WT Mock C WT_dTAG_1 WT dTAG A WT_dTAG_2 WT dTAG B WT_dTAG_3 WT dTAG C KO_Mock_1 KO Mock A KO_Mock_2 KO Mock B KO_Mock_3 KO Mock C KO_dTAG_1 KO dTAG A KO_dTAG_2 KO dTAG B KO_dTAG_3 KO dTAG C

Where the treatment with dTAG makes a conditional KO of my protein. Group A, B and C are different clonal isolations of cells with the inducible KO tag introduced. Therefore the clones will have some differences between each other. The WT should not respond to the dTAG treatment.

In summary:

the controls are: WT mock and dTAG and KO_Mock.

The changes we are interested in only happen in KO_dTAG.

I have tried several designs with ddseq2 and get some hits but I want to make sure I am using the best method.

Code should be placed in three backticks as shown below:

I have played around with the contrasts but it is not clear what the correct contrasts that should be made are.

dds <- DESeqDataSetFromMatrix(countData=round(countsHek), 
                               colData=metaHek, 
                               design = ~ treatment*celltype, tidy = TRUE)
keep <- rowSums(counts(hdds) >= 5) >= 3
dds <- dds[keep,]
dds <- DESeq(dds)
resultsNames(dds)

#[1] "Intercept"                "treatment_Mock_vs_dTAG"   "celltype_WT_vs_KO"        "treatmentMock.celltypeWT"

res <- results(dds, contrast=c(0,1,-1,-1)

# include your problematic code here with any corresponding output 
# please also include the results of running the following in an R session 

sessionInfo( )

I have tried several designs with DESeq2 and get some hits but I want to make sure I am using the best method.

For users who need help with designing their analysis, I recommend to work with a local statistician or someone familiar with linear models in R.

I have to focus my time on the support site to software related questions.