predictModules: Predict gene modules using ICA

In MPCary/DEXICA: Functions for predicting and validating gene modules

Description

Predicts gene modules (sets of co-expressed genes) from a previously preprocessed (see Details) gene expression data matrix using independent component analysis (ICA).

Usage

predictModules(x, n.comp = ceiling(ncol(x) / 4), ...)

Arguments

x	a numeric matrix(-like object), typically with genes in rows and samples (e.g., microarrays) in columns
n.comp	an integer specifying the number of modules to predict
...	additional arguments to be passed to dexFastICA)

Details

This function runs an ICA algorithm (currently only fastICA is supported) on the input matrix. It does not conduct preprocessing operations (such as centering and scaling on the rows and/or columns) on the input matrix first; preprocessMatrix should be run on the input matrix to carryout such steps prior to using this function.

Value

A list with the following elements:

S

The estimated source (gene module definition) matrix

A

The estimated mixing (e.g., weight of each module in each array) matrix

See Also

dexFastICA, preprocessMatrix

Examples

x = matrix(rnorm(100), 10, 10)
x = preprocessMatrix(x)
m = predictModules(x, n.comp = 3)

MPCary/DEXICA documentation built on May 4, 2019, 2:35 p.m.