In your example all of the values are drawn from the same distribution so there will not be substantial differences (row means/variances and column means/variances will be approximately the same).
set.seed(42) d <- matrix(rnorm(100),nrow=20) # Start with your example and modify the row/col means rows <- sample.int(15:25, 20, replace=TRUE) cols <- sample.int(5:15, 5, replace=TRUE) d2 <- sweep(d, 2, cols, "+") d2 <- sweep(d2, 1, rows, "+") heatmap(d2, scale="none") heatmap(d2, scale="row") heatmap(d2, scale="col") ------------------------------------- David L Carlson Associate Professor of Anthropology Texas A&M University College Station, TX 77840-4352 -----Original Message----- From: r-help-boun...@r-project.org [mailto:r-help-boun...@r-project.org] On Behalf Of Witold E Wolski Sent: Wednesday, July 31, 2013 7:04 AM To: r-help@r-project.org Subject: [R] heatmap scale parameter question Would anyone of the more experienced r-users explain to me the behaviour of the scale parameter in the heatmap function. different options for scale (R 3.0.1) do change only the colors but do not affect the dendrograms. Please see for yourself executing the following code: d <- matrix(rnorm(100),nrow=20) stats::heatmap(d) X11() heatmap(d,scale="column") X11() heatmap(d,scale="row") X11() heatmap(d,scale="none") In all four above cases the dendrograms look exactly the same However, scaling clearly affects clustering. see: d <- scale(d) heatmap(d,scale="none") best regards R version 3.0.1 (2013-05-16) -- "Good Sport" ciao -- Witold Eryk Wolski -- Witold Eryk Wolski ______________________________________________ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. ______________________________________________ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
