average.consensus: Consensus network integrated from average edge ranks

In wyguo/RLowPC: Inference of co-expression gene network using relevance low order partial correlation from large scale expression data

Description

Consensus network is built from taking the average ranks of the edges in multiple network predictions.

Usage

average.consensus(adjmatrix.list, directed = F)

Arguments

adjmatrix.list	a list of inferred network matrices with same row and column ordering.
directed	logical. If TRUE, the network is considered as directed. If FALSE, the upper triangular part of the symmetric matrix is used to calculate the rank matrix

Value

average.consensus returns a network with rank average edge ranks. The weights are rescaled to 0-1 and hihger values indicate higher ranks.

Examples

##create two random networks
library(RLowPC)
set.seed(4)
net1<-abs(matrix(rnorm(16),4,4))
net1<-pmax(net1,t(net1))
diag(net1)<-0
set.seed(5)
net2<-abs(matrix(rnorm(16),4,4))
net2<-pmax(net2,t(net2))
diag(net2)<-0
dimnames(net1)<-dimnames(net2)<-list(letters[1:4],letters[1:4])
net.list<-list(net1=net1,net2=net2)
inf.consensus<-average.consensus(adjmatrix.list = net.list,directed = F)
adj2rankadj(net1)
adj2rankadj(net2)
inf.consensus

wyguo/RLowPC documentation built on May 4, 2019, 12:04 p.m.