Dear Experts,

I was provided with 2 groups (normal vs condition) of single channel array gpr files (genepix) that I wished to normalized and analyze for differential expression with the limma package. The arrays are custom for small ncRNA.

Since I'm new at this and have never performed an analysis of this type, I started searching here for help. I found something that is close to what I need to do: https://stat.ethz.ch/pipermail/bioconductor/2013-November/056090.html.

Nevertheless, as Dr. Smyth points out in that post, section 4.5 of Limma's userguide as well as the author of this post both deal with single channel data in the GREEN frequency. I have RED. The option Limma provides for single channel data is the green.only=TRUE but what about if I have only red? 

I've read on https://www.biostars.org/p/8005/ that a possible solution would be to reverse the color settings in the target file, so that I can use the green.only parameter but have it actually pick up the red channel values. But I feel this is weird, shouldn't there be a red.only=TRUE option as well?! I feel this workaround may confuse me afterwards and yield potentially unreliable results later on in the analysis..

Forgive me if the question has already been dealt with in the past (I'm struggling to find where!) but I'm a total newbie.

I'd appreciate any insight/suggestion/reference/help/tutorial of any kind!

~Dario

Hi Dario,

To read the red channel only, here's the work-around:

x <- read.maimages(files, source="genepix", columns=list(G="F635 Mean", Gb="B635 Median"), green.only=TRUE)

The function doesn't have a red.only argument because there didn't seem to be much demand for it -- the green dye (532) is much more common. If using the red dye (635) only is likely to be a regular thing in practice, then I'll add it as an option to read.maimages().