tree.cutting: Optimal cluster number calculation.

In MetChem: Chemical Structural Similarity Analysis

Optimal cluster number calculation.

Description

This function helps to estimate the optimal cluster number that fit the metabolites dataset. It applies different optimal cluster number calculating algorithms to cut clutering tree of hclust function. and return a list contains index corresponding to each cluster number.

Usage

tree.cutting (res,max_nc=20)

Arguments

res	A list produced by KODAMA.chem.sim function.
max_nc	The maximum number of cluster (default = 20).

Value

A list contains the calculation for each clustering of Rousseeuw's Silhouette index.

See Also

KODAMA.chem.sim, WMCSA

Examples

data(Metabolites)

res=KODAMA.chem.sim(Metabolites$SMILES)
clu=tree.cutting(res,max_nc = 30)
plot(clu$min_nc:clu$max_nc,clu$res.S)

MetChem documentation built on June 8, 2025, 10:50 a.m.