[BioC] Limma normalizeBetweenArrays When Samples Have Different Distributions

[Dario Strbenac]

Hello,

I have an experiment where I have ArrayStar Agilent microarrays. Some of them are RNA-IP and some are RNA no antibody control. All are in the green channel. The red channel is not used. One idea is just to use global loess normalisation to get IP / all RNA fold changes, but I also noticed there are probes with names like Negative001 on the array.

Is there a normalisation method available, similar to control normalisation for HumanMethylation450 arrays, that uses these negative control probes ? The method there finds a scaling factor for each array that scales the median of all control probes to a common value, across arrays. Does anyone have experience using control probe normalisation ?

--------------------------------------
Dario Strbenac
Research Assistant
Cancer Epigenetics
Garvan Institute of Medical Research
Darlinghurst NSW 2010
Australia