node_size {mosbi}R Documentation

Node sizes for plotting bicluster networks.

Description

When plotting bicluster networks, node sizes adapted to bicluster sizes can improve visual inspection. Node sizes are computed using the following formula: (atan( (x - min(x)) / (max(x) - min(x)) + offset ) * base_size). With x being defined a vector of bicluster sizes defined by the MARGIN parameter.

Usage

node_size(bics, base_size = 10, offset = 0.2, MARGIN = "column")

Arguments

bics

A list of bicluster objects.

base_size

Is multiplied with the atan result for the node size

offset

Offset for the atan calculation. Has to be > 0. Smaller values result in higher differences of node sizes.

MARGIN

"column", "row" or "both" are taken into account for the size of a bicluster bicluster

Value

Vector of node sizes as floats.

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)
# nz <- node_size(bics)
# plot_algo_network(bn, bics, vertex.size=nz)
# plot(bn, vertex.size=node_size(bics, offset=.1, base_size=15))


[Package mosbi version 1.0.1 Index]