Hi,

I have data from a RNAseq experiment, where different genotypes and treatments were tested. I am analyzing the data using edgeR (fit <- glmFit(dgeObj, design))

The design is as follows, where a, b, c and d are the different factors for the comparison, and 1-8 are the samples. So sample 1,2 and 3 are biological replicates, and so are 5,6, and 7. Factors b and d only have 1 biological replicate.

> design <- data.frame(a=c(1,1,1,0,0,0,0,0), b=c(0,0,0,1,0,0,0,0), c=c(0,0,0,0,1,1,1,0), d=c(0,0,0,0,0,0,0,1))
> design
  a b c d
1 1 0 0 0
2 1 0 0 0
3 1 0 0 0
4 0 1 0 0
5 0 0 1 0
6 0 0 1 0
7 0 0 1 0
8 0 0 0 1

This is the contrast I want to make:

> contrast <- makeContrasts(a - b - (c - d), levels=colnames(design))
> contrast
      Contrasts
Levels (a - b) - (c - d)
     a               1
     b              -1
     c              -1
     d               1

So I want to compare samples that have 3 biological replicates to samples that only have 1. Is this possible?

I don't understand how edgeR gives statistics for the coefficient of interest despite the lack of replicates (how was variance estimated there?). What are your thoughts on this?

Thanks!