Module Identification for multi-layer network

Description

Algorithm for Module Identification on multi-layer network sharing the same set of genes

Usage

1
2
moduleIdentificationGPFixSSMultilayer(W, zi, xrest, x0, lambdai = 1,
  lambda = 1, maxiter = 1000, a = 0.5)

Arguments

W

edge score matrix of the network, n x n matrix

zi

node score vector of this layer, n-length vector

xrest

consensus solution from the rest membership vecconsensus

x0

initial solution, n-length vector

lambdai

parameter in objective, coefficient of node score of current layer

lambda

parameter in objective, coefficient of node score of other layers

maxiter

maximal interation of whole procedure

a

parameter in elastic net the same as in EuclideanProjectionENNORM

Value

a list containing solution for network 1 and network 2

Author(s)

Dong Li, dxl466@cs.bham.ac.uk

References

MOUNTAIN

See Also

vecconsensus

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
n=100
k=20
theta = 0.5
pp <- networkSimulation(n,k,theta)
W <- pp[[1]]
moduleid <- pp[[3]]
z1 <- runif(n, min=0, max=1)
z1[moduleid] <- runif(k, min=theta, max=1)
z2 <- runif(n, min=0, max=1)
z2[moduleid] <- runif(k, min=theta, max=1)
z3 <- runif(n, min=0, max=1)
z3[moduleid] <- runif(k, min=theta, max=1)
x0=rep(1/n,n)
x1=x0
x2=x0
x3=x0
xrest = vecconsensus(cbind(x2,x3))
x1 = moduleIdentificationGPFixSSMultilayer(W,z1,xrest,x0,lambdai=1,lambda=1,maxiter=100,a=0.5)
xrest = vecconsensus(cbind(x1,x3))
x2 = moduleIdentificationGPFixSSMultilayer(W,z2,xrest,x0,lambdai=1,lambda=1,maxiter=100,a=0.5)
xrest = vecconsensus(cbind(x1,x2))
x3 = moduleIdentificationGPFixSSMultilayer(W,z3,xrest,x0,lambdai=1,lambda=1,maxiter=100,a=0.5)