network_edge_strength_float: Count edges in an adjacency matrix using different cut-off...

In tdrose/mosbi: Molecular Signature identification using Biclustering

Count edges in an adjacency matrix using different cut-off thresholds.

Description

Same as network_edge_strength, but for (positive) non-integer matrices.

Usage

network_edge_strength_float(adjm, steps = 100L, maximum = 0)

Arguments

adjm	A symmetrix numeric matrix.
steps	Number of steps for which the edge count is evaluated.
maximum	Highest value until which the edge weight is evaluated. If maximum=0, the max value of adjm is used.

Details

Computes the how many edges remain in a network if edges with a weight lower than a certain threshold are removed. The number of remaining edges between 1 and max(adjm) are calculated. It is assumend that the matrix is symmetric and therefore the number of edges divided by two. Uses the function replace_values_float.

Value

A numeirc matrix of dim(max(adjm), 2). The first column indicated the applied threshold, the second column the remaining edges.

Examples

m <- matrix(seq(1:16), nrow=4)
# m <- matrix(rnorm(10000), nrow=100)
# bics <- c(run_fabia(m), run_isa(m), run_plaid(m))
# bn <- bicluster_network(bics, m)
# network_edge_strength_float(apply_threshold(bn))

tdrose/mosbi documentation built on May 4, 2022, 3:22 p.m.