ReorderToReference: Relabel and reorder clusterings to a reference

In MosaiClusteR: An Umbrella Framework for Multi-Source and Multi-Omics Clustering

Relabel and reorder clusterings to a reference

Description

ReorderToReference relabels and reorders the clusters of a series of clustering results so that the cluster numbers of every method are matched, via a Gale-Shapley stable-matching scheme, to those of the first (reference) method. The result is a matrix in which the columns represent the objects (ordered as for the reference method) and the rows represent the methods. Each cell contains the number of the cluster the object belongs to for that method, relabelled to the reference. It is a possibility that two or more clusters are fused together compared to the reference; in that case the function alerts the user and asks to set fusionsLog to TRUE.

Usage

ReorderToReference(
  List,
  nrclusters = NULL,
  fusionsLog = FALSE,
  weightclust = FALSE,
  names = NULL
)

Arguments

List	A list of clustering outputs to be compared. The first element of the list will be used as the reference.
nrclusters	The number of clusters to cut the dendrogram in. Default is NULL.
fusionsLog	Logical. Indicator for the fusion of clusters. Default is FALSE.
weightclust	Logical. To be used for the outputs of CEC, WeightedClust or WeightedSimClust. If TRUE, only the result of the Clust element is considered. Default is FALSE.
names	Optional. A character vector with the names of the methods.

Value

A matrix of which the cells indicate to what cluster the objects belong to according to the rearranged methods.

Note

The ReorderToReference function was optimized for the situations presented by the data sets at hand. It is noted that the function might fail in a particular situation which results in an infinite loop.

References

Van Moerbeke, M. (2017). MoSaiClusteR: integrative clustering of multi-source data. Hasselt University.

Examples

## Not run: 
data(mosaic_toy)
L <- mosaic_toy$List
res <- list(Cluster(L[[1]], type = "data", distmeasure = "euclidean"),
            Cluster(L[[2]], type = "data", distmeasure = "euclidean"))
M <- ReorderToReference(res, nrclusters = 5, fusionsLog = FALSE,
                        weightclust = FALSE, names = c("S1", "S2"))

## End(Not run)

MosaiClusteR documentation built on Oct. 25, 2026, 1:08 a.m.