Normalize RiboTag samples to input before differential expression with DESeq2

Question

[lozzi]

Hello,

I am trying to perform differential expression analysis on a sample set that looks like this:

  DataFrame with 33 rows and 3 columns
                 ID   Status      Sex
        <character> <factor> <factor>
BP10           BP10  Control   male  
BP17           BP17  Control   female
BP18           BP18  Exp       male  
BP19           BP19  Control   female
BP1             BP1  Input     male  
...             ...      ...      ...
v2_BP5       v2_BP5  Input     female
v2_BP6       v2_BP6  Exp       female
v2_BP7       v2_BP7  Exp       female
v2_BP8       v2_BP8  Exp       female

I want to do differential expression analysis between Control and Experimental with DESeq2 (which I can do) but I want to essentially 'normalize' the samples to the input first, before doing DESeq. Is this possible? I haven't been successful thus far.

Here's what my DESeq code usually looks like if I was just going to compare experimental to control without taking input into consideration:

dds<-DESeqDataSet(se=se,design=~Status)
dds<-DESeq(object=dds)

dds<-estimateSizeFactors(dds)

(res<-results(object=dds,
              alpha=0.05,                      
              lfcThreshold=1.5,                  
              pAdjustMethod='BH',             
              contrast=c('Status','Exp','Control'))) 
summary(res)  

Thanks in advance for any advice!

[Damien]

@lozzi Did you manage to normalize each Ip sample by its input before the differential analysis. I am trying to use the same approach with RiboTag data and DESeq2 as well. Thanks