On Tue, Apr 3, 2012 at 2:41 AM, vinod1 <vinod.hegd...@gmail.com> wrote:
> Sarah,
>
> . clust_tree=hclust(as.dist(x),method="complete")
> . plot(clust_tree)
>
> this produces a dendrogram, whereas
> . clust_tree=hclust(as.dist(x),method="complete")
> . cut = cutree(clust_tree,k=1:5)
> . plot(cut)
>
> produces a plot with 2 dots. The dissimilarity matrix x is 100*100 matrix.
What kind of output do you expect?
If you want to see the clustering tree and an indication of which
object belongs to which cluster, install package WGCNA and use the
function plotDendroAndColors. Continuing with Sarah's example, you can
use
hc <- hclust(dist(USArrests))
labels = cutree(hc, k=1:5)
require("WGCNA")
plotDendroAndColors(hc, labels)
You get the clustering tree plus a color indication of the cluster
each object belongs to in each cut.
You can also produce MDS plots colored by the cluster label.
If you'd like to experiment with more involved ways of identifying
branches (or subtrees) in the dendrogram, I can recommend the article
(warning, shameless plug)
Langfelder P, Zhang B, Horvath S (2007) Defining clusters from a
hierarchical cluster tree: the Dynamic Tree Cut package for R.
Bioinformatics 2008 24(5):719-720
and the package dynamicTreeCut that the short article describes. You
can see some example code at
http://www.genetics.ucla.edu/labs/horvath/CoexpressionNetwork/BranchCutting/
We use this approach in large gene expression data sets.
HTH,
Peter
______________________________________________
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.
