plot.clust: Plots clustering result

In PedCNV: An implementation for association analysis with CNV data.

Description

Makes formatted plots from the clustering result returned from ClusProc.

Usage

  ## S3 method for class 'clust'
 plot(x,
    type = c("histo", "scat", "sil"), adjust = TRUE, ...)

Arguments

x	The clustering results obtained from ClusProc.
type	Factor. For specifying the plot type. It must be one of 'histo', 'scat' and 'sil'. If it is 'histo', the histogram is obtained with the first PC score of the intensity measurement. For 'scat', the first PC score of the intensity measurement is plotted against the mean of the intensity measurement. For 'sil', the silhouette score is plotted. See details.
adjust	Logicals. If TRUE (default), the silhouette-adjusted clustering result will be used. If FALSE, the initial clustering result will be used. See details in ClusProc.
...	Usual arguments passed to the qplot function.

Details

• typeWe provide three types of plots: 'hist', 'scat' and 'sil'. The first two plots are used to visually check the performance of clustering. Different clusters are represented by using different colors. The 'sil' plot is the the overview of the silhouette value for all the individuals, the silhouettes of the different clusters are printed below each other. The higher silhouettes value means the better performance.

Author(s)

Meiling Liu

Examples

# Fit the data under the given clustering numbers
clus.fit <- ClusProc(signal=signal,N=2:6,varSelection='PC.9')
plot(clus.fit,type='histo')

PedCNV documentation built on May 2, 2019, 8:17 a.m.