bayes: Create an adjacency matrix based on score-based structure...

View source: R/statistical.R

bayesR Documentation

Create an adjacency matrix based on score-based structure learning algorithm

Description

'bayes' infers an adjacency matrix using score-based structure learning algorithm 'boot.strength' from the 'bnlearn' package. 'bayes' extracts then the reported connections from running the 'boot.strength' function and assigns the strengths of the arcs of the Bayesian connections to an adjacency matrix. 'bayes' returns this weighted adjacency matrix.

Usage

bayes(x, algorithm = "tabu", R = 100, ...)

Arguments

x

'matrix' where columns are the samples and the rows are features (metabolites), cell entries are intensity values

algorithm

'character', structure learning to be applied to the bootstrap replicates (default is '"tabu"')

R

'numeric', number of bootstrap replicates

...

parameters passed to 'boot.strength'

Details

'boot.strength' measures the strength of the probabilistic relationships by the arcs of a Bayesian network, as learned from bootstrapped data. By default 'bayes' uses the Tabu greedy search.

For use of the parameters used in the 'boot.strength' function, refer to '?bnlearn::boot.strength'. For further information see also Friedman et al. (1999) and Scutari and Nagarajan (2001).

Value

'matrix' with edges inferred from score-based structure learning algorithm 'boot.strength'

Author(s)

Thomas Naake, thomasnaake@googlemail.com

References

Friedman et al. (1999): Data Analysis with Bayesian Networks: A Bootstrap Approach. Proceedings of the 15th Annual Conference on Uncertainty in Artificial Intelligence, 196-201.

Scutari and Nagarajan (2011): On Identifying Significant Edges in Graphical Models. Proceedings of the Workshop Probabilistic Problem Solving in Biomedicine of the 13th Artificial Intelligence in Medicine Conference, 15-27.

Examples

data("x_test", package = "MetNet")
x <- x_test[1:10, 3:ncol(x_test)]
x <- as.matrix(x)
bayes(x, algorithm = "tabu", R = 100)


tnaake/MetNet documentation built on June 30, 2022, 10:50 a.m.