I have microarray data and I want to identify groups of genes that have same expression patterns across my samples. I've done average linkage hierarchical clustering in R. How can I detect which cut-off would be best for dendrogram so I have significant clusters? The height (combination similarity) of my dendrogram goes from 0 to about 2.5 with almost all clusters of genes between the height of 0 and 0.5. How important is it to have a cut-off?
watch here : http://biostar.stackexchange.com/questions/699/