Dear all, I would like to do a principal components analysis of a microarray experiment. In order to do this, I first need to set the expression values of my control spots to NA, so that this analysis is not influenced by these spots. The problem is that I do not know how to achieve this. I can list my control spots by: RG[RG$genes [, "Status"] != "Gene",] I would like to set the corresponding RG$R and RG$G expression values to NA. I can set them to zero by: RG$R <- RG$R * (RG$genes [, "Status"] == "Gene") but this is not what I want. If I try: RG$R <- RG$R * (if(RG$genes [, "Status"] == "Gene") {NA} else {1}) I get the warning message: the condition has length > 1 and only the first element will be used in: if (RG$genes[, "Status"] == "Gene") { Can anybody help me out? Kind regards, Edo Plantinga [[alternative HTML version deleted]]

If you have a RG list, subsetting in limma is pretty easy, I think. See help(subsetting) with limma loaded. To get the subset of your RG list without Control spots, just do (untested): RGnocontrol <- RG[RG$genes[,"Status"]!="Gene",] Sean ----- Original Message ----- From: "Edo Plantinga" <a.e.d.plantinga@med.rug.nl> To: "Bioconductor" <bioconductor@stat.math.ethz.ch> Cc: <stephen@iwinet.rug.nl>; "Rudi Alberts" <r.alberts@cs.rug.nl> Sent: Wednesday, January 28, 2004 4:21 AM Subject: [BioC] Limma: setting expression values to NA > Dear all, > > I would like to do a principal components analysis of a microarray experiment. In order to do this, I first need to set the expression values of my control spots to NA, so that this analysis is not influenced by these spots. The problem is that I do not know how to achieve this. > I can list my control spots by: > RG[RG$genes [, "Status"] != "Gene",] > I would like to set the corresponding RG$R and RG$G expression values to NA. I can set them to zero by: > RG$R <- RG$R * (RG$genes [, "Status"] == "Gene") > but this is not what I want. If I try: > RG$R <- RG$R * (if(RG$genes [, "Status"] == "Gene") {NA} else {1}) > I get the warning message: > the condition has length > 1 and only the first element will be used in: if (RG$genes[, "Status"] == "Gene") { > Can anybody help me out? > > Kind regards, > > Edo Plantinga > > [[alternative HTML version deleted]] > > _______________________________________________ > Bioconductor mailing list > Bioconductor@stat.math.ethz.ch > https://www.stat.math.ethz.ch/mailman/listinfo/bioconductor >

How about something like this? index <- RG$genes[,"Status"]!="Gene" RG$R[index,] <- NA RG$G[index,] <- NA HTH, Jim James W. MacDonald Affymetrix and cDNA Microarray Core University of Michigan Cancer Center 1500 E. Medical Center Drive 7410 CCGC Ann Arbor MI 48109 734-647-5623 >>> "Edo Plantinga" <a.e.d.plantinga@med.rug.nl> 01/28/04 04:21AM >>> Dear all, I would like to do a principal components analysis of a microarray experiment. In order to do this, I first need to set the expression values of my control spots to NA, so that this analysis is not influenced by these spots. The problem is that I do not know how to achieve this. I can list my control spots by: RG[RG$genes [, "Status"] != "Gene",] I would like to set the corresponding RG$R and RG$G expression values to NA. I can set them to zero by: RG$R <- RG$R * (RG$genes [, "Status"] == "Gene") but this is not what I want. If I try: RG$R <- RG$R * (if(RG$genes [, "Status"] == "Gene") {NA} else {1}) I get the warning message: the condition has length > 1 and only the first element will be used in: if (RG$genes[, "Status"] == "Gene") { Can anybody help me out? Kind regards, Edo Plantinga [[alternative HTML version deleted]] _______________________________________________ Bioconductor mailing list Bioconductor@stat.math.ethz.ch https://www.stat.math.ethz.ch/mailman/listinfo/bioconductor

> Dear all, > > I would like to do a principal components analysis of a microarray > experiment. In order to do this, I first need to set the expression > values of my control spots to NA, so that this analysis is not > influenced by these spots. The problem is that I do not know how to > achieve this. I can list my control spots by: > RG[RG$genes [, "Status"] != "Gene",] > I would like to set the corresponding RG$R and RG$G expression values to > NA. I can set them to zero by: RG$R <- RG$R * (RG$genes [, "Status"] == > "Gene") > but this is not what I want. If I try: > RG$R <- RG$R * (if(RG$genes [, "Status"] == "Gene") {NA} else {1}) I get > the warning message: > the condition has length > 1 and only the first element will be used in: > if (RG$genes[, "Status"] == "Gene") { Can anybody help me out? Wouldn't it be better to remove the control spots from your data object entirely? E.g., isGene <- RG$genes$Status =="Gene" RGnocontrols <- RG[isGene,] Cheers Gordon > Kind regards, > > Edo Plantinga