node_size: Node sizes for plotting bicluster networks.

In tdrose/mosbi: Molecular Signature identification using Biclustering

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))

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