Title: Robust Re-Scaling to Better Recover Latent Effects in Data
Version: 1.0
Description: Non-linear transformations of data to better discover latent effects. Applies a sequence of three transformations (1) a Gaussianizing transformation, (2) a Z-score transformation, and (3) an outlier removal transformation. A publication describing the method has the following citation: Gregory J. Hunt, Mark A. Dane, James E. Korkola, Laura M. Heiser & Johann A. Gagnon-Bartsch (2020) "Automatic Transformation and Integration to Improve Visualization and Discovery of Latent Effects in Imaging Data", Journal of Computational and Graphical Statistics, <doi:10.1080/10618600.2020.1741379>.
Date: 2020-5-22
Depends: R (≥ 3.5.0)
License: GPL-3
Encoding: UTF-8
LazyData: true
RoxygenNote: 7.0.2
Imports: DEoptim, nloptr, abind
Suggests: knitr, rmarkdown, testthat, ggplot2, reshape2
VignetteBuilder: knitr
NeedsCompilation: no
Packaged: 2020-05-25 23:56:10 UTC; greg
Author: Gregory Hunt [aut, cre], Johann Gagnon-Bartsch [aut]
Maintainer: Gregory Hunt <ghunt@wm.edu>
Repository: CRAN
Date/Publication: 2020-05-26 11:30:02 UTC

Arc-hyperbolic-sine transformation

Description

Usage

asinh

Format

An object of class list of length 2.

Traditional box-cox power transformation. Accepts one real parameter

Description

Usage

box_cox

Format

An object of class list of length 2.

Exponential of the tranditional box-cox transformation

Description

Usage

box_cox_exp

Format

An object of class list of length 2.

A generalized box-cox transformation that can handle negative data

Description

Usage

box_cox_negative

Format

An object of class list of length 2.

Box-cox transformation with a shift of 1 added to the data

Description

Usage

box_cox_plus1

Format

An object of class list of length 2.

Box-cox transformation with the data shifted so that it is positive

Description

Usage

box_cox_plusmin

Format

An object of class list of length 2.

Box-cox transformation of shifted variable

Description

Usage

box_cox_shift

Format

An object of class list of length 2.

Centers the data column-wise

Description

Centers the data column-wise

Usage

center(x)

Arguments

x

the data.

Calculate the geometric mean

Description

Calculate the geometric mean

Usage

gm_mean(x)

Arguments

x

the data.

Examples

Y <- rlnorm(10)
gm <- gm_mean(Y)

List possible transformations

Description

Returns list of transformations. Each transformation is a transformation function (“T”) accepting a parameter and the derivative of this transformation function (“T_deriv”).

Usage

list_transformations()

Log of the traditional box-cox transformation

Description

Usage

log_box_cox

Format

An object of class list of length 2.

Simple power transformation

Description

Usage

power

Format

An object of class list of length 2.

Re-scale a data matrix

Description

This transformation is three steps (1) Gaussianize the data, (2) z-score Transform the data, and (3) remove extreme outliers from the data. The sequence of these transformations helps focus further analyses on consequential variance in the data rather than having it be focused on variation resulting from the feature's measurement scale or outliers.

Usage

rrscale(
  Y,
  trans_list = list(box_cox_negative = box_cox_negative, asinh = asinh),
  lims_list = list(box_cox_negative = c(-100, 100), asinh = list(0, 100)),
  opt_control = NULL,
  opt_method = "DEoptim",
  z = 4,
  q = 0.001,
  verbose = FALSE,
  log_dir = ".rrscale/",
  zeros = FALSE,
  opts = FALSE,
  seed = NULL
)

Arguments

Y

Data matrix, data.frame, or list of vectors, to be transformed.

trans_list

List of transformations to be considered. See function list_transformations. Each element of the list should be a list containing the transformation function as the first element and the derivative of the transformation function as the second argument. The first argument of each function should be the data, the second the transformation parameter.

lims_list

List of optimization limits for each transformation from trans_list. This should be a list the same length as trans_list. Each element of the list is a two-element vector that sets the optimization limits for the parameter of each transformation family.

opt_control

Optional optimization controlling parameters for DEoptim control argument. See the DEoptim package for details.

opt_method

Which optimization method to use. Defaults to DEoptim. Other choice is nloptr.

z

The O-step cutoff value. Points are removed if their robust z-score is above z in magnitude.

q

The Z-step winsorizing quantile cutoff. The quantile at which to winsorize the data when calculating the robust z-scores.

verbose

a boolean, if TRUE then save optimization output in log_dir.

log_dir

directory for verbose output. Defaults to ".rrscale/"

zeros

How to deal with zeros in the data set. If set to FALSE the algorithm will fail if it encounters a zero. If set to a number or 'NA' then the zeros are replaced by this number or 'NA'.

opts

Boolean determining if optimization output is returned. Defaults to FALSE.

seed

Sets the seed before running any other analyses.

Value

A list of output:

Examples

Y <- rlnorm(10)%*%t(rlnorm(10))
rr.out <- rrscale(Y)
Yt <- rr.out$RR

The completed SVD

Description

This calculates right and left singular vectors of a data matrix possibly containing missing values.

Usage

svdc(X, nu = NULL, nv = NULL)

Arguments

X

the data matrix of which to calcluate the completed SVD.

nu

the number of left singular vectors to calculate

nv

the nubmer of right singular vectors to calculate

Examples

Y <- rnorm(10)%*%t(rnorm(10))
Y[1,1] <- NA
svdc.out <- svdc(Y)

Winsorizes the data

Description

Winsorizes the data

Usage

winsor(x, fraction = 0.01)

Arguments

x

the data.

fraction

the top and bottom quantiles to cap.

Examples

Y <- rlnorm(10)%*%t(rlnorm(10))
Yw <- winsor(Y,1E-2)