**360**wrote:

I am using ROAST. I am a little confused by the results, and I wanted to check if my syntax appears correct. I understand that geneSetTest is using a different method than ROAST. However, I am obtaining substantially different results. Also, I am how does one interpret "Active.Prop=1". Thanks.

# limma syntax

# stage has four levels, and I am interested in the difference between B and C design <- model.matrix(~0+key$stage) colnames(design) = c("A","B","C","D") head(design) # A B C D #1 1 0 0 0 #2 1 0 0 0 #3 1 0 0 0 #4 1 0 0 0 #5 1 0 0 0 #6 1 0 0 0 fit <- lmFit(e, design) fit <- eBayes(fit) cont.matrix <- makeContrasts(risk="B-C",levels=design) fit2 <- contrasts.fit(fit, cont.matrix) fit2 <- eBayes(fit2) tt2 <-topTable(fit2, number=Inf,coef=1) tt2$PROBEID = rownames(tt2) # use geneSetTest # read in probes (not shown) # tt2 is the topTable myIndex = rownames(tt2) %in% probes geneSetTest(myIndex, tt2$t) # 0.2439185 # use roast myIndex = rownames(e) %in% probes roast(e,myIndex,design,contrasts=cont.matrix,nrot=5000)

# the result differs substantially from geneSetTest

# what does it mean that "Active.Prop=1"

# Active.Prop P.Value

#Down 0 1.00000000

#Up 1 0.00019996

#Mixed 1 0.00019996

**37k**• written 4.5 years ago by Juliet Hannah •

**360**