Dear Sir There is a question whether could limma move out outliers, which are not consistently reproducible and has a disproportionately large effect on the statistical result? In my microarray experiment, I do biology repeat and technology repeat, every gene has duplicate in every microarray slide, so I will have 16 data points for each gene. If there is outliers, could I use limma to filter them? Waiting your reply. Best regards. Chen

Filtering weak spots sometimes help. You can filter out or weight spots by for instance the red foreground channel of your RG object: RG$R[(RG$R<200)]<-NA RG$weights[RG$R <200] <- 0.1 I wonder: is it possible to use the fit$stdev.unscaled ? morten > Dear Sir > There is a question whether could limma move out outliers, which are > not consistently reproducible and has a disproportionately large > effect on the statistical result? > In my microarray experiment, I do biology repeat and technology > repeat, every gene has duplicate in every microarray slide, so I will > have 16 data points for each gene. If there is outliers, could I use > limma to filter them? > Waiting your reply. > Best regards. > Chen > >--------------------------------------------------------------------- --- > >_______________________________________________ >Bioconductor mailing list >Bioconductor at stat.math.ethz.ch >https://stat.ethz.ch/mailman/listinfo/bioconductor > >