Re: > > > maDiagnPlots1(test[,2]) > > Error in plot.new() : Figure margins too large Thanks to Robert Gentleman, Joshua Betcher and Sandrine Dudoit for responding. I changed graphics window sizes and output the graphs to files to work around the problem. I also tried to debug the functions to track down exactly where the error is but no luck really - sorry. Next I tried an overall loess normalisation on housekeeping genes exp48.norm<-maNorm(exp48, norm="loess", subset=(housekeep ==1)) and while maPlot(), maDiagnPlots1(), maNormPlots(), work fine (if I'm careful about graphics window sizes), maRawPlots() doesn't > maDiagnPlots1(exp48[,1],fname="test1.ps",dev="postscript",save=T) postscript 2 > maNormPlots(exp48.norm[,1],fname="test2.ps",dev="postscript",save=T) postscript 2 > maRawPlots(exp48[,1],fname="test3.ps",dev="postscript",save=T) Error in factor(z, levels = 1:nlevels(x), labels = levels(x)) : invalid labels; length 0 should be 1 or 2 > Is this a similar error? It sort of looks like it because the boxplot where it stops has no margins Thanks a lot Regards Peter -- Dr Peter Baker, Statistician (Bioinformatics/Genetics), CSIRO Mathematical & Information Sciences, 120 Meiers Rd, INDOOROOPILLY, QLD 4068. Australia. Email: <peter.baker@csiro.au> WWW: http://www.cmis.csiro.au/Peter.Baker/ Phone:+61 7 3214 2210 Fax:+61 7 3214 2881