Hi

I am just trying to work out if this is the correct way of adjusting for having 2 patient groups, disease and non disease, each with different numbers of male/ females. I just want to know what genes are different between the disease and non disease, I don't care about the gender specific genes. This is my code:

design <- model.matrix(~disease+sex)
matrix <- data.matrix(countTable2)
dge <- DGEList(counts=matrix)
dge <- calcNormFactors(dge)
v <- voom(countTable2, design, plot=TRUE)
fit <- lmFit(v,design)
fit <- eBayes(fit)
top2 <- topTable(fit,coef=2,number=Inf,sort.by="P")
sum(top2$adj.P.Val<0.05)

If I include + sex when making the model matrix I get 50 DE genes, and without it I get 46.

Thanks for your help,

Chris

That looks good to me.