文摘
A new radial space-filling method for visualizing cluster hierarchies is presented. The method, referred toas a radial clustergram, arranges the clusters into a series of layers, each representing a different level of thetree. It uses adjacency of nodes instead of links to represent parent-child relationships and allocates sufficientscreen real estate to each node to allow effective visualization of cluster properties through color-coding.Radial clustergrams combine the most appealing features of other cluster visualization techniques but avoidtheir pitfalls. Compared to classical dendrograms and hyperbolic trees, they make much more efficient useof space; compared to treemaps, they are more effective in conveying hierarchical structure and displayingproperties of nodes higher in the tree. A fisheye lens is used to focus on areas of interest, without losingsight of the global context. The utility of the method is demonstrated using examples from the fields ofmolecular diversity and conformational analysis.