cluster_parameters: Detection of algorithm and number of clusters

In OMICsPCA: An R package for quantitative integration and analysis of multiple omics assays from heterogeneous samples

Description

Detection of appropriate clustering algorithm and cluster number for given data using "clvalid" and "NbClust" in background for cluster validation.

Usage

cluster_parameters(name, comparisonAlgorithm = "clValid",
optimal = FALSE, n = 2:6,
clusteringMethods = c("kmeans", "pam"),
validationMethods = c("internal", "stability"),
distance = "euclidean", ...)

Arguments

name	dataframe returned by "extract()".
optimal	logical. If TRUE, returns a dataframe of optimal results
n	a vector of numbers corresponding to the number of clusters to be tested or validated
comparisonAlgorithm	2 choices available: "clValid" or "NbClust"
clusteringMethods	a vector of single or multiple names of clustering algorithms. available choices are: 1) if comparisonAlgorithm = "clValid" : "hierarchical", "kmeans", "diana", "fanny", "som", "model", "sota", "pam", "clara" and "agnes" 2) if comparisonAlgorithm = "NbClust" : "ward.D", "ward.D2", "single", "complete", "average", "mcquitty", "median", "centroid", "kmeans".
validationMethods	name of the method to validate clusters. Available options (one or more): 1) if comparisonAlgorithm = "clValid" then one of: "kl", "ch", "hartigan", "ccc", "scott", "marriot", "trcovw", "tracew", "friedman", "rubin", "cindex", "db", "silhouette", "duda", "pseudot2", "beale", "ratkowsky", "ball", "ptbiserial", "gap", "frey", "mcclain", "gamma", "gplus", "tau", "dunn", "hubert", "sdindex", "dindex", "sdbw", "all" (all indices except GAP, Gamma, Gplus and Tau), "alllong" (all indices with Gap, Gamma, Gplus and Tau included). 2) if comparisonAlgorithm = "NbClust" then one or more of: "internal", "stability", and "biological"
distance	metric used to calculate distance matrix. options: 1) for "clValid" : "euclidean", "correlation", and "manhattan". 2) for "NbClust" : This must be one of: "euclidean", "maximum", "manhattan", "canberra", "binary", "minkowski"
...	additional non-conflicting arguments to "clValid" or "Nbclust"

Value

1) for "clValid"

an object of class "clValid" (optimal = FALSE) or a dataframe of optimal values (optimal = TRUE)

2) for "NbClust"

a list of :

All.index, All.CriticalValues, Best.nc and Best.partition.

See the help pages of "clValid" (?clValid) and "NbClust" (?NbClust) for more details.

Author(s)

Subhadeep Das

References

Brock, G., Pihur, V., Datta, S. and Datta, S. (2008) clValid: An R Package for Cluster Validation Journal of Statistical Software 25(4) http://www.jstatsoft.org/v25/i04

Charrad M., Ghazzali N., Boiteau V., Niknafs A. (2014). "NbClust: An R Package for Determining the Relevant Number of Clusters in a Data Set.", "Journal of Statistical Software, 61(6), 1-36.", "URL http://www.jstatsoft.org/v61/i06/".

Examples

exclude <- list(0,c(1,9))

int_PCA <- integrate_pca(Assays = c("H2az",
"H3k9ac"),
groupinfo = groupinfo,
name = multi_assay, mergetype = 2,
exclude = exclude, graph = FALSE)

name = int_PCA$int_PCA

data <- extract(name = name, PC = c(1:4),
groups = c("WE","RE"), integrated = TRUE, rand = 600,
groupinfo = groupinfo_ext)

#### Using "clValid" ####

clusterstats <- cluster_parameters(name = data,
optimal = FALSE, n = 2:4, comparisonAlgorithm = "clValid",
distance = "euclidean", clusteringMethods = c("kmeans"),
validationMethods = c("internal"))

OMICsPCA documentation built on Nov. 8, 2020, 5:01 p.m.