Dear Gordon, thanks for your answer. There was a problem with the format, I had commas instead of dots for decimal separators. Sorry about that silly mistake, I should have checked all the possible options before posting my email. It now works fine Nevertheless, do you think that using limma for Q-PCR on 96 genes and 19 samples (two conditions) is a valid idea? Maybe it would be more preferable to use some other method instead? David

> Date: Mon, 6 Mar 2006 14:45:13 +0100 (CET) > From: kfbargad at ehu.es > Subject: Re: [BioC] Limma missing values > To: bioconductor at stat.math.ethz.ch > > Dear Gordon, > > thanks for your answer. There was a problem with the format, I had > commas instead of dots for decimal separators. Sorry about that silly > mistake, I should have checked all the possible options before posting > my email. It now works fine > > Nevertheless, do you think that using limma for Q-PCR on 96 genes and > 19 samples (two conditions) is a valid idea? Maybe it would be more > preferable to use some other method instead? In principle, limma should work fine on such data. Best wishes Gordon > David