[BioC] agilent shRNA library

Giusy Della Gatta gd2253 at columbia.edu
Mon Sep 13 22:23:32 CEST 2010


Hi All,

I have microarray data coming from a Agilent custom shRNA library (3 controls "DMSO" and 3 treatments "DBZ").

To normalize these data I used the following code:

library(limma)
targets<-readTargets("targetfile_CycA100nM.txt")
RG<-read.maimages(targets$FileName, source="agilent", ext="txt")
MA<-MA.RG(RG, bc.method="none")
RG.none<-backgroundCorrect (RG,method="none")
MA.n<-normalizeWithinArrays(RG.none, method="loess")
MA.Rq<-normalizeBetweenArrays(MA.n, method="quantile")


By doing the correlation plot between the different samples I noticed that one sample (DBZ#1) looks weird. It seems that there is any correlation between this sample triplicates and between this sample and the control (see attachment). This is happening only with DBZ#1. On the other hand, by using  GeneSpring software from Agilent this sample looks OK.

Please, someone can help me into find out which is the error that I am doing in the script?



Thank you very much,

Giusy



More information about the Bioconductor mailing list