Extracts the Backbone from Networks

An implementation of methods for extracting a sparse unweighted network (i.e. a backbone) from an unweighted network (e.g., Hamann et al., 2016 ), a weighted network (e.g., Serrano et al., 2009 ), or a weighted projection (e.g., Neal et al., 2021 ).

backbone

Welcome

The backbone package implements methods to extract the backbone of a network, which is a sparse and unweighted subgraph that contains only the most "important" or "significant" edges. A backbone can be useful when the original network is too dense, when edge weights are not needed, or when edge weights are difficult to interpret. Methods are available for:

• Weighted networks
• Weighted projections
• Unweighted networks

In addition, the package implements some other utility functions to:

• Randomize matrices while preserving the row and column sums using the fastball algorithm
• Estimate the Bipartite Configuration Model (BiCM)

For more details on these functions and methods, please see:

• vignette("backbone")
• Neal, Z.P. (2026). backbone: An R package to extract network backbones. PLoS ONE, 21, e0349258. https://doi.org/10.1371/journal.pone.0349258
• rbackbone.net

Installation

The /release branch contains the current CRAN release of the backbone package. You can install it from CRAN with:

install.packages("backbone")

The /devel branch contains the working beta version of the next release of the backbone package. All the functions are documented and have undergone various levels of preliminary debugging, so they should mostly work, but there are no guarantees. Feel free to use the devel version (with caution), and let us know if you run into any problems. You can install it You can install from GitHub with:

devtools::install_github("zpneal/backbone", ref = "devel", build_vignettes = TRUE)

Dependencies

The backbone package adopts the tinyverse philosophy, and therefore aims to keep dependencies at a minimum.