I am new to bioconductor and classification algorithms in general. Is there a way to obtain the equation of a decision rule created when using MLInterfaces on a exprSet object? Specifically, I am using stat.diag.daB to implement DLDA. I would like to apply the same decision rule to arrays from multiple datasets. Alternatively, if I define the same arrays as the learning set used as input to MLInterfaces, will it produce the same classification rule? This idea is illustrated using the Golub Expression sets in the code that follows. By my understanding, the classification rules created by the last two lines of code will be the same, but they will be applied to different test sets (arrays 51 through 61 and 62 through 72 respectively). Is this the correct interpretation of how the MLInterfaces functions work? library(golubEsets) data(Golub_Merge) Golub_1<-Golub_Merge[,1:61] Golub_2<-Golub_Merge[,c(1:50,62:72)] dlda.1<-stat.diag.daB(Golub_1,'ALL.AML',1:50) dlda.2<-stat.diag.daB(Golub_2,'ALL.AML',1:50) Thanks, Mike