[BioC] I'm confused about LIMMA statistics

[Sean Davis]

On Mon, Apr 25, 2011 at 6:30 PM, Amy Johnson <a7johnson at gmail.com> wrote:
> Hi,
>
> I'm new in microarray data analysis and I have a quick and maybe "simple"
> question about LIMMA and hope some one can help.
>
> We have 6 Agilent data (3 controls and 3 treated samples). I like to use
> LIMMA to figure out differentially expressing genes. I'm confused about
> LIMMA statistics (logFC, AveExpr, t, P.Value, adj.P.Val, and B). Which one
> researchers typically use to select differentially expressing genes? Can I
> simple use adj.P.Val < 0.05 or P.Value < 0.05? Or, should I use combination
> of these statistics? Thanks.

Hi, Amy.  Your best bet is to thoroughly read the Limma User Guide and
the help pages for ALL the commands you used to generate your topTable
results.  Also, it will help to get some basics of statistics under
your belt.  There are no hard-and-fast rules about what should be
used, but many folks will use adj.P.Val (adjusted to be a False
Discovery Rate) with or without logFC cutoffs.

Sean