Transcriptome trees were constructed by hierarchical clustering of arrays by the average linkage algorithm on the ratio of the relative expression level as estimated by BAGEL in TIGR TMeV 3.1 [ 44].
With a complete enumeration of the pairwise distances between all observations, the linkage algorithm merges the two closest clusters into one, where a cluster can also be a single data point.
As expected, mothur did a better job than ESPRIT when using the average linkage algorithm and the single linkage preclustering significantly improved the performance of the hierarchical clustering.
The analysis was based on raw gene expression values and the 'Average linkage' algorithm, defining the distance between conditions as the average of distances between all pairs of individuals in all groups.
More recently, Corander et al. (2004) applied the complete linkage algorithm to the output form their MCMC sampler (the software package BAPS).
Additionally, a hierarchical clustering was applied to the protein spots showing quantitative patterns of expression using the Euclidean distance and the average linkage algorithm by the Cluster 3.0 and Java Treeview software.
A hierarchical clustering was applied on the transcripts with significant quantitative differences in expression, using the uncentered correlation distance based on the Pearson correlation and the average linkage algorithm by the Cluster 3.0 and Java treeview software http://bonsai.ims.u-tokyo.ac.jp/~mdehoon/software/cluster/software.htm.htm
P450 genes were then clustered using the complete linkage algorithm of the OC cluster analysis program [ 12] based on the sequence identity of all pairwise genes reported from SSEARCH.
Figure 3 shows the rooted binary forest generated by applying the exact linkage algorithm to the pooled sample of 60 000 observations.
Hierarchical clustering of samples based on their predicted probability values of oncogenic pathway activation was performed using the complete linkage algorithm with the Euclidean distance metric [37].
The exact linkage algorithm takes the posterior co-assignment probabilities as input, and yields as output a rooted binary tree, - or more generally, a forest of such trees.
