Swap edges of a bipartite graph

Run the edge swapping algorithm from Milo et al. (2003) on a bipartite graph while maintaining the separation of the nodes in two groups.

bipartite_edge_swap(gr, Q = 20, N = Q * igraph::ecount(gr))

Arguments

gr	a tidygraph object with a node attribute called type that holds boolean values (i.e. either TRUE or FALSE)
Q	number of permutations you are conducting; default is 20
N	the number of edge swaps; default is \(Q \times |E(G)|\)

Value

the graph with \(N\) random edge swaps with constrained marginals

Examples

set.seed(0)
bgr <- tidygraph::create_ring(6, directed = FALSE)
bgr <- tidygraph::mutate(bgr, type = rep(c(TRUE, FALSE), 3))
print(bgr)
#> # A tbl_graph: 6 nodes and 6 edges
#> #
#> # A bipartite simple graph with 1 component
#> #
#> # Node Data: 6 x 1 (active)
#>   type 
#>   <lgl>
#> 1 TRUE 
#> 2 FALSE
#> 3 TRUE 
#> 4 FALSE
#> 5 TRUE 
#> 6 FALSE
#> #
#> # Edge Data: 6 x 2
#>    from    to
#>   <int> <int>
#> 1     1     2
#> 2     2     3
#> 3     3     4
#> # … with 3 more rows
print(bipartite_edge_swap(bgr, 10))
#> # A tbl_graph: 6 nodes and 6 edges
#> #
#> # A bipartite simple graph with 1 component
#> #
#> # Edge Data: 6 x 2 (active)
#>    from    to
#>   <int> <int>
#> 1     1     6
#> 2     4     5
#> 3     1     2
#> 4     3     4
#> 5     5     6
#> 6     2     3
#> #
#> # Node Data: 6 x 2
#>   type   .idx
#>   <lgl> <int>
#> 1 TRUE      1
#> 2 FALSE     2
#> 3 TRUE      3
#> # … with 3 more rows