Hypergraphs are generalisations of graphs. Graphs connect together vertices in a binary manner, that is that (in a simple graph) each edge will connect two vertices. In a hypergraph, there are still vertices like there are in a graph, but edges are replaced with hyperedges that can connect together any number of vertices. This allows us to represent higher order interactions that are useful in the analysis of many systems.
rhype allows the user to create hypergraph objects in R from different
data structures. It then contains many functions for transforming the
hypergraph or calculating properties of hypergraphs that may lead to
deeper insights of the data that it represents.
rhype is available from CRAN using
install.packages("rhype")or the development version is available from GitHub using
remotes::install_github("hwarden162/rhype")rhype is currently undergoing a large rewrite. The main focus of this
is to complete unit testing for all functions to cover unoriented
hypergraphs with real coefficients. As this happens there will be bug
fixes as well as updates to the way in which rhype stores hypergraph
objects.
Excitingly, on top of this the development version of rhype
(v0.3.0.9000 currently avaliable on GitHub) now has functions for
finding support graphs and connectivity graphs. These can be used
together to find shortest hyperpaths between vertices.