\HeaderA{08.Tests}{Hypothesis Testing for Linear Models}{08.Tests}
\keyword{documentation}{08.Tests}
\begin{Description}\relax
LIMMA provides a number of functions for multiple testing across both contrasts and genes.
The starting point is an \code{MArrayLM} object, called \code{fit} say, resulting from fitting a linear model and running \code{eBayes} and, optionally, \code{contrasts.fit}.
See \LinkA{06.LinearModels}{06.LinearModels} or \LinkA{07.SingleChannel}{07.SingleChannel} for details.
\end{Description}
\begin{Section}{Multiple testing across genes and contrasts}
The key function is \code{\LinkA{decideTests}{decideTests}}.
This function writes an object of class \code{\LinkA{TestResults}{TestResults}}, which is basically a matrix of \code{-1}, \code{0} or \code{1} elements, of the same dimension as \code{fit\$coefficients}, indicating whether each coefficient is significantly different from zero.
A number of different multiple testing strategies are provided.
The function calls other functions \code{\LinkA{classifyTestsF}{classifyTestsF}}, \code{\LinkA{classifyTestsP}{classifyTestsP}} and \code{\LinkA{classifyTestsT}{classifyTestsT}} which implement particular strategies. 
The function \code{\LinkA{FStat}{FStat}} provides an alternative interface to \code{classifyTestsF} to extract only the overall moderated F-statistic.

A number of other functions are provided to display the results of \code{decideTests}.
The functions \code{\LinkA{heatDiagram}{heatDiagram}} (or the older version \code{\LinkA{heatdiagram}{heatdiagram}} displays the results in a heat-map style display.
This allows visual comparison of the results across many different conditions in the linear model.

The functions \code{\LinkA{vennCounts}{vennCounts}} and \code{\LinkA{vennDiagram}{vennDiagram}} provide Venn diagram style summaries of the results.

Summary and \code{show} method exists for objects of class \code{TestResults}.

The results from \code{decideTests} can also be included when the results of a linear model fit are written to a file using \code{\LinkA{write.fit}{write.fit}}.
\end{Section}
\begin{Section}{Other Functions}
Given a set of p-values, the function \code{\LinkA{convest}{convest}} can be used to estimate the proportion of true null hypotheses.

When evaluating test procedures with simulated or known results, the utility function \code{\LinkA{auROC}{auROC}} can be used to compute the area under the Receiver Operating Curve for the test results for a given probe.
\end{Section}
\begin{Author}\relax
Gordon Smyth
\end{Author}