Hi, I am using DESeq2 to look at the effect of different diet over time. I have 3 different diet (D1, D2, D3), and 3 time points (T1, T2, T3). I am trying to create best model for this, therefore I tried

dds <- DESeqDataSetFromMatrix(countData=rawCounts_totat, 
                                          colData=mapping_file, 
                                          design=~Diet+Time+Diet:Time)
dds.1 <- DESeq(dds, parallel = T)
resultsNames(dds.1)
[1] "Intercept"       "Diet_D1_vs_ctr" "Diet_D2_vs_ctr" "Diet_D3_vs_ctr" "Time_T2_vs_T1"   "Time_T3_vs_T1"  
[7] "DietD1.TimeT2"  "DietD2.TimeT2"  "DietD3.TimeT2"  "DietD1.TimeT3"  "DietD2.TimeT3"  "DietD3.TimeT3" 

table(colData(dds.1)$Time, colData(dds.1)$Diet)
# ctr here means control    
     ctr  D1  D2  D3
  T1  12  12  12  12
  T2  12  12  12  12
  T3  12  12  12  12

Question 1: But my question is to compare within Diet 1 (D1) over time, that means I need to look at the comparison between T1 vs T2, T1 vs T3, and T2 vs T3 for Diet 1, similarly for other Diets (D2 and D3).

Question 2: Secondly, what does lets say, "DietD1.TimeT2" mean here?

res.D1 <- results(dds.1, contrast=c("Diet", "D1", "ctr"))

Question 3: If I use the above contrast function to compare between Diet 1 and control (ctr), does this uses all the samples present in Diet 1 (12+12+12) and in the control (12+12+12)?

But I want to use contrast function to compare between Diet 1 (n=12) and control (n=12) at T1 or at T2 or at T3.

Any suggestion would be helpful.

Many thanks!

If you don't understand what your code is doing, I strongly suggest you put down your data, and run through one or two tutorials with sample data first. (Hint #1; don't just throw interactions into your design. Omit them unless you are sure that you need them and that you are sure you understand how they change the results you can get out)