# compareModules: compareModules In hferg/EMMLiv2: A Maximum Likelihood Approach to the Analysis of Modularity

## Description

This takes a correlation matrix or 3D landmark array, a model definition, and then two module numbers or names to compare. It plots (if plot = TRUE) a figure of three boxplots - the first two are the correltions within each of the two modules, and the third is the between-module correlations. These boxes are coloured such that matching colours are not significantly different according to a Tukey HSD test. The results of the anova and tukey HSD test are also returned.

## Usage

 `1` ```compareModules(corr, model, test_modules, plot = TRUE) ```

## Arguments

 `corr` A correlation matrix or a 3D array of landmarks. If 3D then a correlation matrix is calculated with dotcorr `model` Either a vector of numbers describing a model of modules, or a 2 column dataframe with the first bein landmark names and the second being the module definitions. `test_modules` A vector of two module numbers to compare, or if the modules are named, the names of those two modules. `plot` Logical - if TRUE the plot is drawn.

## Value

A list with two elements - the first is the result of an ANOVA compaing the mean correlations within- and between-modules, and the second is the results of a TukeyHSD test on that ANOVA. If plot = TRUE a plot showing these results is called.

## Examples

 ```1 2 3 4 5 6 7 8 9``` ```data(macacaCorrel) data(macacaModels) # Pick a model to draw modules from - as a vector. model <- macacaModels\$Goswami compareModules(corr = macacaCorrel, model = model, test_modules = c(2, 5)) # Or as a 2 column dataframe... model <- macacaModels[ , c(1, 4)] compareModules(corr = macacaCorrel, model = model, test_modules = c(2, 5)) ```

hferg/EMMLiv2 documentation built on Nov. 30, 2017, 4:35 p.m.