contrasts.varFit: Compute contrasts for a varFit object.

Description

Compute estimated coefficients, standard errors and LogVarRatios for a given set of contrasts.

Usage

contrasts.varFit(fit, contrasts=NULL)

Arguments

fit

list containing a linear model fit produced by varFit. The fit object should be of class MArrayLM.

contrasts

numeric matrix with rows corresponding to coefficients in fit and columns containing contrasts.

Value

A list object of the same class as fit.

Details

This function calls the contrasts.fit function in limma to compute coefficients and standard errors for the specified contrasts corresponding to a linear model fit obtained from the varFit function. LogVarRatios are also computed in terms of the contrasts. A contrasts matrix can be computed using the makeContrasts function.

Examples

Run this code

# Randomly generate data for a 3 group problem with 100 CpG sites and 4 arrays in each group. 

library(limma)

y<-matrix(rnorm(1200),ncol=12)

group<-factor(rep(c(1,2,3),each=4))
design<-model.matrix(~0+group)
colnames(design)<-c("grp1","grp2","grp3")

# Fit linear model for differential variability
vfit<-varFit(y,design)

# Specify contrasts
contr<-makeContrasts(grp2-grp1,grp3-grp1,grp3-grp2,levels=colnames(design))

# Compute contrasts from fit object
vfit.contr<-contrasts.varFit(vfit,contrasts=contr)

summary(decideTests(vfit.contr))

# Look at top table of results for first contrast

topVar(vfit.contr,coef=1)