Hi Jim, I am using the pdmclass package and had a question. I'd like to be able to plot the output from pdmClass when there are only 2 levels. However, that generates an error. Here is the example from the pdf documentation with a change to show the problem: library("pdmclass") data("fibroEset") y <- as.factor(pData(fibroEset)[, 2]) x <- t(exprs(fibroEset)) gn.class <- pdmClass(y ~ x, method = "pls") plot(gn.class, pch = levels(y)) Now change the levels of y so there are only two: y <- as.character(pData(fibroEset)[, 2]) > y [1] "b" "b" "b" "b" "b" "b" "b" "b" "b" "b" "b" "g" "g" "g" "g" "g" "g" "g" "g" "g" "g" "g" "g" "h" "h" "h" "h" "h" "h" "h" "h" "h" "h" "h" "h" "h" [37] "h" "h" "h" "h" "h" "h" "h" "h" "h" "h" y[y=="g"] <- "b" y <- as.factor(y) y [1] b b b b b b b b b b b b b b b b b b b b b b b h h h h h h h h h h h h h h h h h h h h h h h Levels: b h gn.class <- pdmClass(y ~ x, method = "pls") plot(gn.class, pch = levels(y)) Error in rbind(vars[, coord.pair], means[, coord.pair]) : subscript out of bounds predict(gn.class) [1] b b b b b b b b b b b b b b b b b b b b b b b h h h h h h h h h h h h h h h h h h h h h h h Levels: b h I've poked around the mda:::plot.fda function and I can see why I get an error. But I was wondering, given predict still works with 2 levels, how to plot the predictions? Thanks much, Dick ********************************************************************** ********* Richard P. Beyer, Ph.D. University of Washington Tel.:(206) 616 7378 Env. & Occ. Health Sci. , Box 354695 Fax: (206) 685 4696 4225 Roosevelt Way NE, # 100 Seattle, WA 98105-6099 http://depts.washington.edu/ceeh/members_fc_bioinfo.html http://staff.washington.edu/~dbeyer ***