gen_simple: Generate Simulated Data Sets from a Simple Model

In eLNNpaired: Model-Based Gene Clustering for Genomics Data from Paired/Matched Designs

Description

Generate Simulated Data Sets from a Simple Model.

Usage

gen_simple(
  G, 
  n = 30, 
  psi = c(0.441, 1, -0.442, 1, 2), 
  t_pi = c(0.086, 0.071)
)

Arguments

G	An integer, the number of genes.
n	An integer, the number of pairs for each gene.
psi	A vector of 5 elements containing model parameters mu_1, sigma_1, mu_2, sigma_2, and sigma_3.
t_pi	the cluster proportion for cluster 1 (over-expressed probes) and cluster 2 (under-expressed probes).

Details

We assume there are three clusters of gene probes: (1) over-expressed; (2) under-expressed; and (3) non-differentially expressed. For probes in cluster 1, we assume the within-pair log2 difference of gene expression is from N(mu_1, sigma_1^2). For probes in cluster 2, we assume the within-pair log2 difference of gene expression is from N(mu_2, sigma_2^2). For probes in cluster 3, we assume the within-pair log2 difference of gene expression is from N(0, sigma_3^2). mu_1>0 and mu_2<0.

Value

An ExpressionSet object, the feature data frame of which include memGenes.true (3-cluster membership for gene probes) and memGenes2.true (2-cluster membership for gene probes).

In 3-cluster membership, 1 indicates over-expressed, 2 indicates under-expressed, and 3 indicates non-differentially expressed.

In 2-cluster membership, 1 indicates differentially expressed, 0 indicates non-differentially expressed.

Author(s)

Yunfeng Li <[e-mail]> and Weiliang Qiu <[e-mail]>

References

Li Y, Morrow J, Raby B, Tantisira K, Weiss ST, Huang W, Qiu W. (2017), <doi:10.1371/journal.pone.0174602>

Examples

es=gen_simple(
  G = 500,
  n = 30,
  psi = c(0.441, 1, -0.442, 1, 2),
  t_pi = c(0.086, 0.071)
)

print(es)

eLNNpaired documentation built on May 29, 2017, 12:04 p.m.