progenyClust - Finding the Optimal Cluster Number Using Progeny Clustering
Implementing the Progeny Clustering algorithm, the
'progenyClust' package assesses the clustering stability and
identifies the optimal clustering number for a given data
matrix. It uses k-means clustering as a default, provides a
tailored hierarchical clustering function, and can be
customized to work with other clustering algorithms and
different parameter settings. The package includes a main
function progenyClust(), plot and summary methods for
'progenyClust' object, a function hclust.progenyClust() for
hierarchical clustering, and two example datasets (test and
cell) for testing.