backbone: backbone: Extracts the Backbone from Graphs

backboneR Documentation

backbone: Extracts the Backbone from Graphs


Provides methods for extracting from an unweighted and sparse subgraph (i.e., a backbone) that contains only the most "important" edges in a weighted bipartite projection, a non-projection weighted network, or an unweighted network.

Available backbone extraction functions include:

  • For weighted bipartite projections of weighted bipartite networks: osdsm().

  • For weighted bipartite projections of binary bipartite networks: fixedfill(), fixedrow(), fixedcol(), sdsm(), and fdsm().

  • For non-projection weighted networks: global(), disparity().

  • For unweighted networks: sparsify(), sparsify.with.skeleton(), sparsify.with.gspar(), sparsify.with.lspar(), sparsify.with.simmelian(), sparsify.with.jaccard(), sparsify.with.meetmin(), sparsify.with.geometric(), sparsify.with.hypergeometric(), sparsify.with.localdegree(), sparsify.with.quadrilateral().

  • For all networks: backbone.suggest() will examine the data and suggest an appropriate backbone function

The package also includes some utility functions:

  • fastball() - Fast marginal-preserving randomization of binary matrices

  • bicm() - Compute probabilities under the bipartite configuration model

For additional documentation and background on the package functions, see vignette("backbone"). For updates, papers, presentations, and other backbone news, please see


Neal, Z. P. (2022). backbone: An R Package to Extract Network Backbones. PLOS ONE, 17, e0269137. doi: 10.1371/journal.pone.0269137

backbone documentation built on Oct. 19, 2022, 1:09 a.m.