createData: Simulate Data

In FamilyRank: Algorithm for Ranking Predictors Using Graphical Domain Knowledge

Description

Simulate data sets meant to emulate gene expression data in oncology.

Usage

createData(n.case, n.control, mean.upper = 13, mean.lower = 5, 
sd.upper = 1, sd.lower = 1, n.features = 10000, 
subtype1.feats = 1:5, subtype2.feats = 6:10, subtype3.feats = 11:15)

Arguments

n.case	Number of cases to simulate.
n.control	Number of controls to simulate
mean.upper	Mean of upper component of bimodal Gaussian distribution from which features are simulated.
mean.lower	Mean of lower component of bimodal Gaussian distribution from which features are simulated.
sd.upper	Standard deviation of upper component of bimodal Gaussian distribution from which features are simulated.
sd.lower	Standard deviation of lower component of bimodal Gaussian distribution from which features are simulated.
n.features	Number of features to simulate
subtype1.feats	Index of features used to define subtype 1.
subtype2.feats	Index of features used to define subtype 2.
subtype3.feats	Index of features used to define subtype 3.

Details

Simulates case/control data as described in createCase and createControl, and graphical domain knowledge as described in createGraph.

Value

Returns a named list with a simulated feature matrix (x), simulated binary response vector (y), vector of subtype labels (subtype), and simulated domain knowledge graph (graph).

Author(s)

Michelle Saul

References

ADD REFERENCE

See Also

createCase, createControl, createGraph

Examples

## Toy Example
# Simulate data set
# 10 samples
# 20 features
# Features 1 through 15 perfectly define response
# All other features are random noise.
data <- createData(n.case = 5, n.control = 5, mean.upper=13, mean.lower=5,
                   sd.upper=1, sd.lower=1, n.features = 20,
                   subtype1.feats = 1:5, subtype2.feats = 6:10,
                   subtype3.feats = 11:15)
x <- data$x
y <- data$y
graph <- data$graph

FamilyRank documentation built on Feb. 5, 2021, 9:05 a.m.