plotClustersPCA {DIRECT} R Documentation

## PCA Plot for Posterior Allocation Probability Matrix

### Description

Function plotClustersPCA generates a Principal Components Analysis (PCA) plot for the posterior mean estimate of allocation probability matrix. The first two principal components are used. See figures in Fu, Russell, Bray and Tavare.

### Usage

plotClustersPCA(item.names, data.summary,


### Arguments

 item.names A vector of character strings, indicating how each item should be labeled in the PCA plot. data.summary The list generated from summaryDIRECT that contains processed posterior estimates. PCA.label.adj A scalar to be added to the coordinates of item.names for better display. ... Additional arguments for plot.

### Details

The PCA plot produced here displays the uncertainty in the inferred clustering. Each inferred cluster is shown with a distinct color. The closer two clusters are in the PCA plot, the higher the level of uncertainty in inferring these two clusters.

None.

Audrey Q. Fu

### References

Fu, A. Q., Russell, S., Bray, S. and Tavare, S. (2013) Bayesian clustering of replicated time-course gene expression data with weak signals. The Annals of Applied Statistics. 7(3) 1334-1361.

summaryDIRECT for processing MCMC estimates for clustering and generating the list data.summary used here.
plotClustersMean, plotClustersSD, plotSimulation.
## See example in DIRECT.