data.gen: Simulated Data Generation

In nlnet: Nonlinear Network, Clustering, and Variable Selection Based on DCOL

Description

Generating gene matrix as a example of input.

Usage

 
data.gen(n.genes=100, n.samples=100, n.grps=10, aver.grp.size=10, 
n.fun.types=6, epsilon=0.1, n.depend=0)

Arguments

n.genes	the number of rows of the matrix.
n.samples	the number of columns of the matrix.
n.grps	the number of hidden clusters.
aver.grp.size	averge number of genes in a cluster.
n.fun.types	number of function types to use.
epsilon	noise level.
n.depend	data generation dependence structure. can be 0, 1, 2.

Details

The data generation scheme is described in detail in IEEE ACM Trans. Comput. Biol. Bioinform. 10(4):1080-85.

Value

return the data including gene and clustering.

data	the gene matrix
grps	the predicted clustering

Author(s)

Tianwei Yu<[e-mail]>

Examples

##generating a gene matrix with 100 genes, some in 5 clusters, and 100 samples per gene.
output<-data.gen(n.genes=100, n.samples=10, n.grps=5)
##get the gene matrix from the source of data.
matrix<-output$data
##get the hiden clusters from the source of data.
grps<-output$grp

nlnet documentation built on Jan. 13, 2021, 10:35 a.m.