Optimal Leaf Ordering for Hierarchical Clustering

In this page we present in detail our results on optimal leaf ordering for hierarchical clustering. This page accompanies a paper (which has the same title) that was submitted to ismb01. In that paper we introduced our optimal ordering algorithm, analyzed its time and space complexity and suggested improvements to the algorithm that make its running time very reasonable. In addition we presented several examples that show how optimal ordering helps in analyzing and interpreting data. In this page we present these examples and other examples which we did not present in the paper due to lack of space. In addition, for our biological results, the next pages contain links to our complete ordered results (that is, the actual excel sheets of the gene ordering).


Random data results:
 
Increase in similarity as a function of the number of leaves

Hand generated results:
 
Input 1 Input 2

Large scale results:
Input 1 Input 2 Input 3


Biological results:
Cell cycle (24 cdc15 experiments)
Cell cycle (59 experiments, using cdc15, cdc28 and alpha factor)
Complexes clustering results