adj.remove.cycles: Function to remove cycles that may be present in a directed...

In predictionet: Inference for predictive networks designed for (but not limited to) genomic data

Description

This function removes cycles that may be present in a directed graph represented by an adjacency matrix,

Usage

adj.remove.cycles(adjmat, from, maxlength)

Arguments

adjmat	adjacency matrix with positive entries represent evidence for the presence of an edge and entries less or equal than zero represent absence of an edge; parents in row, children in columns.
from	indices or names of nodes for which the cycles present in the childhood should be removed; if missing, all cycles will be removed.
maxlength	maximum length of path, once this length is reached no longer paths will be searched for.

Details

This function may be useful when it comes to generate a bayesian network using a topology identified from an source of information where cycles are allowed. When cycles are removed, the function tries to keep the most positive entries.

Value

A list of two items

adjmat.acyclic	an adjacency matrix without cycles
adjmat.removed	a matrix of booleans representing the edges that have been removed from the original adjacency matrix to make it acyclic

Author(s)

Benjamin Haibe-Kains

Examples

set.seed(54321)
xx <- matrix(sample(c(0,1), 100, replace=TRUE), nrow=10, ncol=10)
adj.remove.cycles(adjmat=xx, from=1, maxlength=3)

Example output

Loading required package: igraph

Attaching package: 'igraph'

The following objects are masked from 'package:stats':

    decompose, spectrum

The following object is masked from 'package:base':

    union

Loading required package: catnet
$adjmat.acyclic
    X1 X2 X3 X4 X5 X6 X7 X8 X9 X10
X1   0  0  0  1  1  1  0  0  0   0
X2   0  0  0  0  0  1  1  1  1   1
X3   0  0  0  1  1  0  0  1  1   0
X4   0  1  0  0  0  0  1  1  1   1
X5   0  1  1  1  0  1  0  1  0   1
X6   1  1  1  0  1  0  1  0  1   1
X7   0  1  1  1  1  0  0  1  1   1
X8   0  0  1  1  1  0  0  0  1   0
X9   0  1  0  1  0  0  0  0  0   1
X10  0  0  1  1  1  0  1  0  0   0

$adjmat.removed
       X1    X2    X3    X4    X5    X6    X7    X8    X9   X10
X1  FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
X2  FALSE  TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
X3  FALSE FALSE  TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
X4  FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
X5  FALSE FALSE FALSE FALSE  TRUE FALSE FALSE FALSE FALSE FALSE
X6  FALSE FALSE FALSE FALSE FALSE  TRUE FALSE FALSE FALSE FALSE
X7  FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
X8  FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
X9  FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE  TRUE FALSE
X10 FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE  TRUE

predictionet documentation built on Nov. 8, 2020, 7:48 p.m.