I am trying to use the EBcoexpress package to do differential coexpression analysis. In it's documentation they say it requires the expression matrix and condition array which are straightforward, and a pattern object created by coexpression which I have no idea what it represents and their explanation was not clear.

Their code that I am mostly copying is:

library(EBcoexpress)
data(fiftyGenes) # A 50-by-125 expression matrix
tinyCond <- c(rep(1,100),rep(2,25))
tinyPat <- ebPatterns(c("1,1","1,2"))
initHP <- initializeHP(D, tinyCond)
zout <- ebCoexpressZeroStep(D, tinyCond, tinyPat, initHP)

I can't use this exactly as it i as my data has three conditions while their condition matrix has 2 groups. Does anyone know how can I adapt ebPatterns for my own purposes, or what is the exact purpose of that function here.