bnclustNetworks: Deriving consensus networks based on posterior probabilities...

In bnClustOmics: Bayesian Network-Based Clustering of Multi-Omics Data

Deriving consensus networks based on posterior probabilities of mixture model

Description

This function derives consensus models of networks representing all clusters based on several threshold for posterior probabilities of individual edges.

Usage

bnclustNetworks(
  bnres,
  bnnames,
  sump = 1.2,
  minp = 0.5,
  minkp = 0.9,
  maxkp = NULL
)

Arguments

bnres	an object of class 'bnclustOmics'; see bnclustOmics
bnnames	an object of class 'bnInfo'; see bnInfo
sump	threshold for the sum of posterior probabilities in all discovered networks
minp	threshold for the minimum posterior probability in at least one network, when the sum of posterioirs is bigger than sump
minkp	threshold for the minimum posterior probability in at least one network, when the sum of posterioirs is less than sump
maxkp	(optional) threshold for the maximum posterior probability in at least one network; used to esclude cluster specific edges from the edges with high sum of posterioirs (>sump)

Value

returns a list of adjacency matrices, one for each cluster representing consensus models

Examples

bnnames<-bnInfo(simdata,c("b","c"),c("M","T"))
intlist<-bnclustNetworks(bnres3,bnnames)

bnClustOmics documentation built on Aug. 5, 2022, 5:11 p.m.