Learn R Programming

BipartiteModularityMaximization (version 1.23.120.1)

bipmod: Partition bipartite network into non-overlapping biclusters, by optimizing bipartite modularity.

Description

This function partitions a bipartite network into non-overlapping biclusters by optimizing bipartite modularity defined in Barber (2007) using the bipartite version of the algorithm described in Treviño (2015).

Usage

bipmod(incid_mat, ITER = 10)

Value

MODULARITY Modularity value (Q).

ASSIGN Integer labels representing partition of rows followed by columns in same order as incidence matrix.

Arguments

incid_mat

Incidence matrix of a bipartite network.

ITER

A positive integer representing the number of iterations used to maximizing modularity, (default=10).

Details

The function takes as input a bipartite network represented as an incidence matrix (using a matrix or a data frame) with non-negative values (the row sums and column sums must be positive, to ensure there are no disconnected nodes). The function partitions the rows and columns into non-overlapping submatrices (biclusters), and outputs the membership of rows and columns to a partition, and modularity (Q) representing the quality of the partitioning.

References

Barber, M. J. (2007). Modularity and community detection in bipartite networks. Physical Review E, 76(6), 066102. <doi:10.1103/PhysRevE.76.066102>

Trevino, S., Nyberg, A., Del Genio, C. I., & Bassler, K. E. (2015). Fast and accurate determination of modularity and its effect size. Journal of Statistical Mechanics: Theory and Experiment, 2015(2), P02003. <doi:10.1088/1742-5468/2015/02/P02003>

Examples

Run this code
data(example_data)
bipmod(example_data)

Run the code above in your browser using DataLab