pcp: Projection of points into one dimension.

In jasonserviss/ClusterSignificance: The ClusterSignificance package provides tools to assess if class clusters in dimensionality reduced data representations have a separation different from permuted data

Description

Project points onto a principal curve.

Usage

getData(x, ...)

## S4 method for signature 'Pcp'
getData(x, n = NULL)

## S4 method for signature 'Pcp'
initialize(.Object, ..., classes, points.orig, line,
  points.onedim, index, class.color)

## S4 method for signature 'Pcp,missing'
plot(x, y, steps = "all", class.color = NULL, ...)

pcp(mat, ...)

## S4 method for signature 'matrix'
pcp(mat, classes, df = NULL, warn = TRUE,
  class.color = NULL, ...)

## S4 method for signature 'Pcp'
show(object)

Arguments

x	matrix object for the function pcp otherwise it is a Pcp object
...	additional arguments to pass on
n	data to extract from Pcp (NULL gives all)
.Object	internal object
classes	vector in same order as rows in matrix
points.orig	multidimensional points describing the original data
line	multidimensional points describing a line
points.onedim	a vector of points
index	internal index from the projection
class.color	user assigned group coloring scheme
y	default plot param, which should be set to NULL
steps	1,2,3,4,5,6 or "all"
mat	matrix with samples on rows, PCs in columns. Ordered PCs, with PC1 to the left.
df	degrees of freedom, passed to smooth.spline
warn	logical indicating if a change in the default df argument should generate a warning. mostly for internal use.
object	Pcp object

Details

The resulting Pcp object containing results from a principal curve reduction to one dimension. The group and the one dimensional points will be the information needed to carry out a classification using the classify() function. As a help to illustrate the details of the dimension reduction, the information from some critical steps is stored in the object. To visually explore these there is a dedicated plot method for Pcp objects, use plot().

Value

The pcp function returns an object of class Pcp

Author(s)

Jesper R. Gadin and Jason T. Serviss

Examples

#use demo data
data(pcpMatrix)
classes <- rownames(pcpMatrix)

#run function
prj <- pcp(pcpMatrix, classes)

#getData accessor
getData(prj)

#getData accessor specific
getData(prj, "line")

#plot the result (if dim >2, then plot in 3d)
plot(prj)

#plot the result (if dim=2, then plot in 2d)
prj2 <- pcp(pcpMatrix[,1:2], classes)
plot(prj2)

jasonserviss/ClusterSignificance documentation built on May 9, 2019, 5:56 p.m.