Thresher (version 1.1.5)

SimThresher-class: Class "SimThresher"

Description

The SimThresher class is used to simulate Thresher objects.

Usage

SimThresher(ss, nSample, nm = deparse(substitute(ss)), rho = NULL,
            agfun = agDimTwiceMean, ...)

Value

The SimThresher function returns an object of the

SimThresher class.

Objects from the Class

Objects should be created using the SimThresher constructor.

Arguments

ss

A covariance matrix.

nSample

An integer; the number of samples to simulate.

nm

A character string; the name of this object.

rho

A numeric vector; the correlation between different variables. If NULL, then these are obtained from the covariance matrix.

agfun

A function used by the AuerGervini function to determine the number of principal components.

...

Parameters to be passed to the Thresher constructor.

Slots

nSample:

An integer; the number of simulated samples.

covariance:

A covariance matrix.

rho:

A vector of correlation coefficients; essentially the unique values in the upper triangular part of the covariance matrix.

Extends

Class Thresher, directly.

Methods

image

signature(x = "SimThresher"): Produces an image of the covariance matrix.

makeFigures

signature(object = "SimThresher"): This is a convenience function to produce a standard set of figures. In addition tot he plots preodcued forThresher object, this function also produces an image of te covariance matrix used in the simulations. If the DIR argument is non-null, it is treated as the name of an existing directory where the figures are stored as PNG files. Otherwise, the figures are displayed interactively, one at a time, in a window on screen.

Author

Kevin R. Coombes <[e-mail]>, Min Wang.

Details

Basically, given a number of samples and a covariance matrix, simulate a data matrix of the appropriate size and multivariate normal structure by assuming that all of the means are zero. After simulating the data, we apply the Thresher algorithm. The result is an object that combines the simulation parameters, simulated data, and fitted model.

References

Wang M, Abrams ZB, Kornblau SM, Coombes KR. Thresher: determining the number of clusters while removing outliers. BMC Bioinformatics, 2018; 19(1):1-9. doi://10.1186/s12859-017-1998-9.

See Also

Thresher, Reaper.

Examples

set.seed(250264)
rho <- 0.5
nProtein <- 16
splinter <- sample((nProtein/2) + (-3:3), 1)
sigma1 <- matrix(rho, ncol=nProtein, nrow=nProtein)
diag(sigma1) <- 1
st <- SimThresher(sigma1, nSample=300)
image(st, col=redgreen(64), zlim=c(-1,1))
screeplot(st, col='pink', lcol='red')
plot(st)
scatter(st)
heat(st)