average.consensus: Consensus network integrated from average edge ranks

Description Usage Arguments Value Examples

View source: R/average.consensus.R

Description

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

Usage

1
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

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
##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 28, 2017, 7:34 a.m.