HCsnip-package: Semi-supervised adaptive-height snipping of the Hierarchical...

In HCsnip: Semi-supervised adaptive-height snipping of the Hierarchical Clustering tree

Description

This package contain functions for extracting meaningful clusters from a HC tree. Rather than cutting the HC tree at a fixed highest (as existing methods do), it snips the tree at variable heights to extract hidden clusters. Cluster extraction process uses both the data type from which HC tree is derived and the available patients follow-up information. Functions for testing the significance of extracted clusters and cluster visualization using sample's molecular entropy are also given. If two HC trees are presented which maybe corresponding to the two treatment groups, this package also includes functions for optimally assigning new patients to one of the two HC trees and calculate the expected gain in terms of follow-up.

Details

Package:	HCsnip
Type:	Package
License:	GPL (>= 2)

Author(s)

Askar Obulkasim Maintainer: Askar Obulkasim <[e-mail]>

References

Obulkasim,A. et al., (2013). "Semi-supervised adaptive-height snipping of the Hierarchical Clustering tree", submitted.

HCsnip documentation built on Nov. 17, 2017, 11:17 a.m.