overlap: Calculate overlaps between every pair of biclusters.

In jonalim/mfBiclust: Pipelines for Bi-Clustering Using Matrix Factorization

Description

The overlap between biclusters i and j is defined as the number of matrix elements shared by the two biclusters divided by the size of bicluster i.

Usage

overlap(BiclusterRows, BiclusterCols, matrix = FALSE)

Arguments

BiclusterRows	a list, each element containing row indices in the bicluster
BiclusterCols	a list, each element containing column indices in the bicluster
matrix	if true, returns a matrix A where element A_{i,j} is the overlap between biclusters i and j. Otherwise returns a list where the ith element is a vector giving the overlap between bicluster i and each bicluster, including itself. In both instances, the order of biclusters is preserved from the input.

Value

If matrix, a numeric matrix. Else, a list of numeric vectors.

Examples

bce <- BiclusterExperiment(yeast_benchmark[[1]])
bce <- addStrat(bce, k = 2, method = "als-nmf")
bcs <- getStrat(bce, 1)
biclusterLists <- biclusterMatrix2List(clusteredFeatures(bcs),
 clusteredSamples(bcs))
biclusterRows <- biclusterLists[[1]]
biclusterCols <- biclusterLists[[2]]
overlap(biclusterRows, biclusterCols)
overlap(biclusterRows, biclusterCols, matrix = TRUE)

jonalim/mfBiclust documentation built on May 4, 2019, 4:13 a.m.