caiwei at mdanderson.org wrote: > John, > > Thanks, > > What does this command do (in the user guide)? does it affect Log2Ratio? > > RG$design<-c(-1,-1) > A quick glance at the code for log2ratios() reveals that the design component of either the MAList or SegList is not accessed when returning the results, so it does not correct for dye swaps. However, if one looks at the code for the run...() functions, the first few lines DO use the $design component, so any calculations done for the segmentation do correct for dye swaps. It is probably worth a quick glance at the code to satisfy yourself that you know what is going on. You can see code by typing the function without any arguments or parentheses: log2ratios runDNAcopy etc. Hope that helps, Sean

It adds the design component to your RGList/MAlist object. You would either use: > RG$design <- c(-1,-1) #Cy5 as ref channel or > RG$design <- c(1,1) #Cy3 as ref channel This is required before you use the processCGH function Regards John Quoting caiwei at mdanderson.org on Mon 08 Oct 2007 23:48:38 BST: > > John, > > Thanks, > > What does this command do (in the user guide)? does it affect > Log2Ratio? > > RG$design<-c(-1,-1) > > Caimiao > > > > |---------+------------------------------------------> > | | | > | | | > | | jhs1jjm at leeds.ac.uk| > | | | > | | | > | | | > | | 10/08/2007 05:28 PM| > | | | > | | | > | | | > |---------+------------------------------------------> > > >--------------------------------------------------------------------- -------------------------------| > | > | > | > | > | > | > |To: > | > | caiwei at mdanderson.org > | > |cc: > | > | "bioconductor at stat.math.ethz.ch" > <bioconductor at="" stat.math.ethz.ch=""> | > | > | > | > | > | > | > | > | > |Subject: > | > | Re: [BioC] snapCGH: log2ratio > | > | > | > > >--------------------------------------------------------------------- -------------------------------| > > > > > Hi Caimiao, > > It is the log2(Cy5/Cy3) i.e log2(red/green). Depending on what you > read > into limma with read.maimages this will be the default estimates > referred to in the vignette for your particular array or whatever you > specified using the columns argument. Whether this is log(sample/ref) > or log(ref/sample) depends on which way round you labeled the samples > I > guess. In aCGH expts the ref channel is normally dyed using Cy5 and > test channel dyed using Cy3. > > Regards > > John > > > Quoting caiwei at mdanderson.org on Mon 08 Oct 2007 19:24:20 BST: > > > > > Dear list, > > > > Are the Log2 ratios in our MAlist the log ratios of test > > sample/reference > > sample? or the log ratios of Cy5/Cy3 as when we run > > normalizeWithinArrays > > in Limma? I am confused when I read the section of "Reading Data" > > about > > the design vector . > > > > Thanks, > > > > Caimiao > > > > _______________________________________________ > > Bioconductor mailing list > > Bioconductor at stat.math.ethz.ch > > https://stat.ethz.ch/mailman/listinfo/bioconductor > > Search the archives: > > http://news.gmane.org/gmane.science.biology.informatics.conductor > > > > > > > >