Hi All,

I am trying to generate a design for my RNA-seq data and I am not able to get the right comparison. This is my sampleTable:

sampleName         condition       time
Mock12hRep1       Mock12h         12
Mock12hRep2       Mock12h         12
Mock12hRep3       Mock12h         12
Mock3hRep1         Mock3h             3
Mock3hRep2         Mock3h             3
Mock3hRep3         Mock3h             3
Treated12hRep1   Treated12h      12
Treated12hRep2   Treated12h      12
Treated12hRep3   Treated12h      12
Treated3hRep1     Treated3h         3
Treated3hRep2     Treated3h         3
Treated3hRep3     Treated3h         3
UntreatedRep1     Untreated          0
UntreatedRep2     Untreated          0
UntreatedRep3     Untreated          0

The comparison I want to do is:

Treated3h_vs_Mock3h
Treated12h_vs_Mock12h 
Treated_vs_Mock   
Mock_vs_Untreated

If I use condition as design

condition
 [1] Mock12h    Mock12h    Mock12h    Mock3h     Mock3h     Mock3h     Treated12h Treated12h Treated12h Treated3h  Treated3h 
[12] Treated3h  Untreated  Untreated  Untreated 
Levels: Mock12h Mock3h Treated12h Treated3h Untreated

I get this

resultsNames(dds)
[1] "Intercept"                       "condition_Mock3h_vs_Mock12h"     "condition_Treated12h_vs_Mock12h"
[4] "condition_Treated3h_vs_Mock12h"  "condition_Untreated_vs_Mock12h" 

Please if anyone can help me figure the right design for my experiment.

Any help would be appreciated! Thanks!

There’s no problem here. Use contrast to compare the levels of interest. See the vignette or ?results for an example.