PSEA (version 1.6.0)

fmm: Full model matrix

Description

Generates a model matrix containing (quantitative) independent variables and interaction regressors coding for group-specific effects for each of tthe quantitative variable.

Usage

fmm(v,d,int=TRUE)

Arguments

v

n-by-m numeric matrix. Matrix containing m quantitative variables as columns (i.e. n is the number of samples).

d

n-by-o numeric matrix. Matrix containing o binary regressors (encoding o group-specific differences) as columns. n is the number of samples.

int

logical. If TRUE (default) an intercept is included as the first column of the full model matrix.

Value

fmm

numeric matrix. The full model matrix. Its dimension is is n-by-(m*(o+1)+1) if int is TRUE and n-by-(m*(o+1)) if int is FALSE.

Details

Quantitative variables are specified via the argument "v". Sample groups are specified using binary (i.e. 0,1) variables (argument "d"). The returned model matrix represents a linear model with "m" quantitative variables and "o" group effects (corresponding to "o"+1 sample groups).

Specifically, the columns of the full model matrix contain (in this order): an intercept (if "int" is TRUE), the quantitative variables in the same order as they are provided in the input, group-specific effects for each quantitative variable (encoded as interaction regressors and in the same order as the quantitative variable are provided in the input).

An interaction regressor corresponds a quantitative variable multiplied by a (group-specific) binary variable. It encodes the difference (for the particular variable) between a specific group and the reference group.

Examples

## Load example expression data (variable "expression")
## for 23 transcripts and 41 samples, and associated
## phenotype (i.e. group) information (variable "groups")
data("example")

## The group data is encoded as a binary vector where
## 0s represent control samples (first 29 samples) and
## 1s represent disease samples (last 12 samples)
groups

## Four cell population-specific reference signals
## (i.e. quantitative variable)
neuron_probesets <- list(c("221805_at", "221801_x_at", "221916_at"),
                "201313_at", "210040_at", "205737_at", "210432_s_at")
neuron_reference <- marker(expression, neuron_probesets)

astro_probesets <- list("203540_at",c("210068_s_at","210906_x_at"),
                "201667_at")
astro_reference <- marker(expression, astro_probesets)

oligo_probesets <- list(c("211836_s_at","214650_x_at"),"216617_s_at",
                "207659_s_at",c("207323_s_at","209072_at"))
oligo_reference <- marker(expression, oligo_probesets)

micro_probesets <- list("204192_at", "203416_at")
micro_reference <- marker(expression, micro_probesets)

## Full model matrix with an intercept, 4 quantitative variables and
## group-specific (disease vs control) differences for the
## 4 quantitative variables
fmm(cbind(neuron_reference, astro_reference, oligo_reference, 
	micro_reference), groups)