LModularity {evolqg}

L Modularity

Description

Calculates the L-Modularity (Newman-type modularity) and the partition of traits that minimizes L-Modularity. Wrapper for using correlations matrices in community detection algorithms from igraph.

Usage

LModularity(cor.matrix, method = optimal.community, ...)

Arguments

cor.matrix

correlation matrix

method

community detection function

...

Additional arguments to igraph community detection function

Details

Warning: Using modularity maximization is almost always a terrible idea. See: https://skewed.de/tiago/blog/modularity-harmful

Value

List with L-Modularity value and trait partition

Note

Community detection is done by transforming the correlation matrix into a weighted graph and using community detection algorithms on this graph. Default method is optimal but slow. See igraph documentation for other options.

If negative correlations are present, the square of the correlation matrix is used as weights.

References

Modularity and community structure in networks (2006) M. E. J. Newman, 8577-8582, doi: 10.1073/pnas.0601602103

Examples

## Not run: 
# A modular matrix:
modules = matrix(c(rep(c(1, 0, 0), each = 5),
rep(c(0, 1, 0), each = 5),
rep(c(0, 0, 1), each = 5)), 15)
cor.hypot = CreateHypotMatrix(modules)[[4]]
hypot.mask = matrix(as.logical(cor.hypot), 15, 15)
mod.cor = matrix(NA, 15, 15)
mod.cor[ hypot.mask] = runif(length(mod.cor[ hypot.mask]), 0.8, 0.9) # within-modules
mod.cor[!hypot.mask] = runif(length(mod.cor[!hypot.mask]), 0.3, 0.4) # between-modules
diag(mod.cor) = 1
mod.cor = (mod.cor + t(mod.cor))/2 # correlation matrices should be symmetric

# requires a custom igraph installation with GLPK installed in the system
LModularity(mod.cor)
## End(Not run)