clusters-methods: create a grouping based on either the raw data or a MDS...

In stela2502/BioData: The package allows to annotate a numeric data.table with col and row data

Description

This function clusters the data based on hclust, mclust or the kmeans algorithm. This function can be used to cluster any data slot the BioData class offers (see combination of onwhat and clusterby).

Usage

clusters(
  dataObj,
  clusterby = "raw",
  useGrouping = NULL,
  groups.n = 3,
  ctype = "hierarchical clust",
  onwhat = "Expression",
  cmethod = "ward.D2",
  name = NULL
)

Arguments

dataObj	the BioData object
clusterby	is used to specify the MDS type if onwhat is set to MDS (PCA, TSNE_R ...) or the raw, zscored or raw slot for expression data (default 'raw')
useGrouping	do nothing and simply use this grouping default=NULL
groups.n	how many groups should be detected default= 3
ctype	cluster type - either 'hierarchical clust', 'mclust' or 'kmeans' default = 'hierarchical clust'
onwhat	This option selects the source data ('Expression', or 'MDS')
cmethod	the method to used with the hclust clustering (default = 'ward.D2')
name	the name for the new grouping (default = 'auto_clusters.1:n')

Examples

clusters( TestData , clusterby = "TSNE_R", groups.n = 3, ctype = "kmeans", onwhat= 'MDS', name = "kmeansTSNE_R clusters" )

stela2502/BioData documentation built on Feb. 23, 2022, 5:47 a.m.