radial.ranking: Radial Ranking of MST

Description Usage Arguments Details Value Author(s) References See Also Examples

View source: R/radial.ranking.R

Description

Rank vertices in an object of class igraph (see package igraph for the definition of class igraph) that consists of a minimum spanning tree (MST) or the union of multiple MSTs radially such that vertices with higher depth and distance from the centroid are given higher ranks.

Usage

1

Arguments

object

object of class igraph that consists of a minimum spanning tree or the union of multiple spanning trees.

Details

Rank nodes in an object of class igraph (see package igraph) that consists of a minimum spanning tree (MST) or the union of multiple MSTs radially. The MST is rooted at the node of smallest geodesic distance (centroid) and nodes with largest depths from the root are assigned higher ranks. Hence, ranks are increasing radially from the root of the MST (Friedman and Rafsky 1979).

Value

Numeric vector giving the radial node ranks in the MST or union of MSTs.

Author(s)

Yasir Rahmatallah and Galina Glazko

References

Rahmatallah Y., Emmert-Streib F. and Glazko G. (2012) Gene set analysis for self-contained tests: complex null and specific alternative hypotheses. Bioinformatics 28, 3073–3080.

Friedman J. and Rafsky L. (1979) Multivariate generalization of the Wald-Wolfowitz and Smirnov two-sample tests. Ann. Stat. 7, 697–717.

See Also

HDP.ranking, RKStest, RMDtest.

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
## generate random data using normal distribution
## generate 20 features in 20 samples
object <- matrix(rnorm(400),20,20)
objt <- aperm(object, c(2,1))
## calculate the weight matrix
Wmat <- as.matrix(dist(objt, method = "euclidean", diag = TRUE, upper = TRUE, p = 2))
## create a weighted undirectional graph from the weight matrix
gr <- graph_from_adjacency_matrix(Wmat, weighted = TRUE, mode = "undirected")
## find the minimum spanning tree
MST <- mst(gr)
radial.ranks <- radial.ranking(MST)

GSAR documentation built on Nov. 8, 2020, 7:16 p.m.