Dear R users,

I have a project that needs to use linear model of limma to do the contrast analysis, but the difficult is to use numeric variable in the contrast. We would like to use Batch as random effect, Stage as numeric variable and Age as co-variate in analysis. But I don't know how to set the contrast using Stage (numeric variable with 0,1,2,3 on 4 different stages). The comparison will be 1+2+3 (represent the post-infection) to 0 (pre-infection), I don't know if there is a way to solve this in limma or other package.

Thanks a lot and appreciate it for any input.

Best,

Li

From your description, your experiment should be straightforward to analyse using limma. However I think there is some confusion about numeric covariates vs factors.

If you want to form a contrast of Stage 1+2+3 vs 0, then Stage must be declared as a factor rather than simply as a numeric variable.

I am guessing that if you simply declare Stage <- factor(Stage), and then redo the analysis, that the difficulties you may be having will go away.