cataloguesClustering: Clustering of Catalogues or Signatures

In Nik-Zainal-Group/signature.tools.lib: signature.tools.lib

Clustering of Catalogues or Signatures

Description

Cluster sample catalogues or signatures using hierarchical clustering with average linkage. The distance metric used is 1 - cosine similarity, so the clustering is based on the shape/direction of the catalogue/signature vectors.

Usage

cataloguesClustering(samplescatalogues, nclusters, outdir = NULL)

Arguments

samplescatalogues	catalogues or signatures, with colnames as sample or signature names and rownames as the channel names
nclusters	number of clusters, can be a single positive integer or a list of integers
outdir	if specified, plot results into the outdir folder

Value

returns a list containing clustering results for each number of clusters requested

Examples

resClust <- cataloguesClustering(catalogues,2:10,"results/")

Nik-Zainal-Group/signature.tools.lib documentation built on April 13, 2025, 5:50 p.m.