BANFF (version 1.1)

EM.HODC: Hierarchical Ordered Density Clustering (HODC) for Finite Mixture Model Fitting

Description

This function implements the HODC algorithm for density estimates by a finite mixture of normals model which is implemented by function Mclust in R package mclust.

Usage

EM.HODC(pvalue)

Arguments

pvalue

a vector of p-values obtained from large scale statistical hypothesis testing

Value

a list of density clustering results by the HODC algorithm

mean

mean estimates of unimportant and important clusters

variance

variance estimates of unimportant and important clusters

pro

probability estimates of unimportant and important clusters

classification

classifications configurations

Details

This function calls Mclust to estimate the marginal density of the testing statistics r, converted from p-values, using a mixture of normal densities without incorporating the network information. Furthermore, it implements the HODC algorithm to classify density components into two clusters. We refer to them as the unimportant cluster and the important cluster, where the important cluster has a larger mean than the unimportant cluster.

References

Yize Zhao, Jian Kang, Tianwei Yu (2014) A Bayesian non parametric model for selecting gene and gene sub network, Annals of Applied Statistics, in press.

Zhou Lan, Jian Kang, Tianwei Yu, Yize Zhao, BANFF: an R package for network identifications via Bayesian nonparametric mixture models, working paper.

Examples

rstat=c(rnorm(50,mean=1),rnorm(50,mean=2),rnorm(100,mean=4),rnorm(100,mean=8))
pvalue=pnorm(-rstat)
EMHODC=EM.HODC(pvalue)