Fig. 3
From: Genome-scale MicroRNA target prediction through clustering with Dirichlet process mixture model
Clusters resulting from cascade DPGMM model. (a) Clustering tree shown at bottom left. A section of the tree is enlarged at the top right. Ellipses are homogeneous clusters. Numbers in parenthesis represent number of examples in the cluster. Striped boxes are mixed clusters, width of each stripe corresponds to the percentage of a miRNA in the mixture. (b) (top) the distribution of number of examples in clusters over all clusters; (bottom) distribution of percentages of dominant miRNAs in each cluster. A dominant miRNA has highest presence in a cluster