[BioC] DESeq Design issue
Simon Anders
anders at embl.de
Tue Feb 5 19:06:46 CET 2013
Hi Joshua
On 05/02/13 18:57, Joshua Pollack [guest] wrote:
>> condition <- c(rep("Control",3), rep("Rods",3), rep("Cubes",3))
>> batch <- c(rep(c(1,2,3),3))
>> Design <- as.data.frame(cbind(condition,batch),row.names=colnames(data))
>> CDS <- newCountDataSet(data,Design)
Instead of the last line, try
CDS <- newCountDataSet(data,condition)
Otherwise, DESeq thinks that 'batch' is a further factor in which your
samples differ and won't believe anymore that the replicates are replicates.
Simon
More information about the Bioconductor
mailing list