Hi

I know variations on this question have been asked and I have read them, but I'd like to verify what I am doing is correct. I want to find genes that are differentially expressed between any of my samples at once as one would do in an ANOVA.

This is what I normally do, to find those DE between two different treatments:

colnames(design) <- c('V1', 'V2', 'V4', 'V6', 'V8', 'SERO') # sero is a covariate
contrast.matrix <- makeContrasts('V1-V2', 'V1-V4', 'V1-V6', 'V1-V8', levels = design)

matrix <- data.matrix(data2)
fit <- lmFit(matrix,design)
fit2 <- contrasts.fit(fit, contrast.matrix)
fit2 <- eBayes(fit2)

top2 <- topTable(fit2,coef=4,number=Inf,sort.by="P")
sig <- subset(top2, P.Value<0.05)
nrow(sig)

This is what I have done to get DE genes between any condition. I am not sure if this is correct.

anova <- topTableF(fit2, number=Inf)
anova_sig <- subset(anova, P.Value <= 0.05)

I guess I will have to correct these p values for multiple testing.

Thanks again for your help.

That's pretty much correct. Alternatively, you can run topTable without specifying coef, which will give you the same results as running topTableF. And yes, you should be working with the BH-adjusted p-values (i.e., adj.P.Val) when considering which genes are DE.