Help for package o2plsda

Type:

Package

Title:

Multiomics Data Integration

Version:

0.0.25

Description:

Provides functions to do 'O2PLS-DA' analysis for multiple omics data integration. The algorithm came from "O2-PLS, a two-block (X±Y) latent variable regression (LVR) method with an integral OSC filter" which published by Johan Trygg and Svante Wold at 2003 <doi:10.1002/cem.775>. 'O2PLS' is a bidirectional multivariate regression method that aims to separate the covariance between two data sets (it was recently extended to multiple data sets) (Löfstedt and Trygg, 2011 <doi:10.1002/cem.1388>; Löfstedt et al., 2012 <doi:10.1016/j.aca.2013.06.026>) from the systematic sources of variance being specific for each data set separately.

License:

GPL-3

Imports:

Rcpp (≥ 1.0.7), dplyr, magrittr, parallel, ggplot2, ggrepel, methods, stats

Encoding:

UTF-8

Suggests:

knitr, markdown, rmarkdown

VignetteBuilder:

knitr

LinkingTo:

Rcpp, RcppArmadillo

RoxygenNote:

7.3.1

NeedsCompilation:

yes

Repository:

CRAN

Packaged:

2024-07-18 20:02:16 UTC; bioguo

Author:

Kai Guo [aut, cre], Junguk Hur [aut], Eva Feldman [aut]

Maintainer:

Kai Guo <guokai8@gmail.com>

Date/Publication:

2024-07-18 22:40:02 UTC

do cross-validation with group factors

Description

do cross-validation with group factors

Usage

.o2cv(X, Y, n, nx, ny, group, nr)

Arguments

X

a Numeric matrix (input)

Y

a Numeric matrix (input)

n

Integer. Number of joint PLS components.

nx

Integer. Number of orthogonal components in X

ny

Integer. Number of orthogonal components in Y

group

a vector to indicate the group for Y

nr

Integer to indicate the folds for cross validation

Value

a list of o2pls results

Author(s)

Kai Guo

Partial least squares discriminant analysis

Description

Perform a PLS discriminant analysis

Usage

.pls(X, Y, nc, cv = TRUE, nr_folds = 5)

Arguments

X

a matrix of predictor variables.

Y

a single vector indicate the group

nc

the number of pls components .

cv

logical indicating whether cross-validation will be performed or not (suggest TRUE).

nr_folds

nr_folds Integer to indicate the folds for cross validation. @return list with PLS results

trans matrix * matrix

Description

trans matrix * matrix

Usage

AtA(A)

Value

A matrix

Class "O2pls" This class represents the Annotation information

Description

Class "O2pls" This class represents the Annotation information

Slots

X: a Numeric matrix (input)
Y: a Numeric matrix (input)
params: paramaters ysed in o2pls analysis
results: list of o2pls results

Author(s)

Kai Guo

calcualte the Q value

Description

calcualte the Q value

Usage

Q(y, y_hat)

Value

a numeric

get the sd for a matrix

Description

get the sd for a matrix

Usage

colsds(X)

Value

standard deviation of the matrix

get the total values for a matrix

Description

get the total values for a matrix

Usage

column_sums(X)

Value

sum value of the column

two matrix mutiplication

Description

two matrix mutiplication

Usage

eigenmult(A, B)

Value

A matrix

three matrix mutiplication

Description

three matrix mutiplication

Usage

eigenthree(A, B, C)

Value

A matrix

MCCV sampling

Description

MCCV sampling

Usage

getMCCV_cpp(x, n)

Value

a vector with random sampling based on group-balanced Monte Carlo cross-validation

Extract the loadings from an O2PLS fit

Description

This function extracts loading parameters from an O2PLS fit

Usage

loadings(x, ...)

## S3 method for class 'O2pls'
loadings(x, loading = c("Xjoint", "Yjoint", "Xorth", "Yorth"), ...)

Arguments

x

Object of class O2pls

...

For consistency

loading

the loadings for one of "Xjoint", "Yjoint", "Xorth", "Yorth"

Value

Loading matrix

extract the loading value from the O2PLSDA analysis

Description

extract the loading value from the O2PLSDA analysis

Usage

## S3 method for class 'o2plsda'
loadings(x, loading = "Xloading", ...)

Arguments

x

Object of class o2plsda

loading

the loadings for one of "Xjoint", "Yjoint", "Xorth", "Yorth"

...

For consistency

extract the loading value from the PLSDA analysis

Description

extract the loading value from the PLSDA analysis

Usage

## S3 method for class 'plsda'
loadings(x, ...)

Arguments

x

Object of class plsda

...

For consistency

Cross validation for O2PLS

Description

Cross validation for O2PLS

Usage

o2cv(
  X,
  Y,
  nc,
  nx,
  ny,
  group = NULL,
  nr_folds = 5,
  ncores = 1,
  scale = FALSE,
  center = FALSE
)

Arguments

X

a Numeric matrix (input)

Y

a Numeric matrix (input)

nc

Integer. Number of joint PLS components.

nx

Integer. Number of orthogonal components in X

ny

Integer. Number of orthogonal components in Y

group

a vector to indicate the group for Y

nr_folds

Integer to indicate the folds for cross validation

ncores

Integer. Number of CPUs to use for cross validation

scale

boolean values determining if data should be scaled or not

center

boolean values determining if data should be centered or not

Value

a data frame with the Q and RMSE values

Author(s)

Kai Guo

Examples

set.seed(123)
X = matrix(rnorm(500),50,10)
Y = matrix(rnorm(500),50,10)
X = scale(X, scale = TRUE)
Y = scale(Y, scale = TRUE)
# group factor could be omitted if you don't have any group 
group <- rep(c("Ctrl","Treat"), each = 25)
cv <- o2cv(X, Y, 1:2, 1:2, 1:2, group=group, nr_folds = 2, ncores=1)

fit O2PLS model with best nc, nx, ny

Description

fit O2PLS model with best nc, nx, ny

Usage

o2pls(X, Y, nc, nx, ny, scale = FALSE, center = FALSE)

Arguments

X

a Numeric matrix (input)

Y

a Numeric matrix (input)

nc

Integer. Number of joint PLS components.

nx

Integer. Number of orthogonal components in X

ny

Integer. Number of orthogonal components in Y

scale

boolean values determining if data should be scaled or not

center

boolean values determining if data should be centered or not

Value

An object containing

Xscore

Joint X scores

Xloading

Joint X loadings

Yscore

Joint Y scores

Yloading

Joint Y loadings

TYosc

Orthogonal X scores

PYosc

Orthogonal X loadings

WYosc

Orthogonal X weights

UXosc

Orthogonal Y scores

PXosc

Orthogonal Y loadings

CXosc

Orthogonal Y weights

BU

Regression coefficient in Tt ~ U

BT

Regression coefficient in U ~ Tt

Xhat

Prediction of X with Y

Yhat

Prediction of Y with X

R2Xhat

Variation of the predicted X as proportion of variation in X

R2Yhat

Variation of the predicted Y as proportion of variation in Y

R2X

Variation of the modeled part in X (defined by Joint + Orthogonal variation) as proportion of total variation in X

R2Y

Variation of the modeled part in Y (defined by Joint + Orthogonal variation) as proportion of total variation in Y

R2Xcorr

Variation of the joint part in X

R2Ycorr

Variation of the joint part in Y

R2Xo

Variation of the orthogonal part in X as proportion of variation in X

R2Yo

Variation of the orthogonal part in Y as proportion of variation in Y

R2Xp

Variation in X joint part predicted by Y Joint part

R2Yp

Variation in Y joint part predicted by X Joint part

varXj

Variation in each Latent Variable (LV) in X Joint part

varYj

Variation in each Latent Variable (LV) in Y Joint part

varXorth

Variation in each Latent Variable (LV) in X Orthogonal part

varYorth

Variation in each Latent Variable (LV) in Y Orthogonal part

Exy

Residuals in X

Fxy

Residuals in Y

Author(s)

Kai Guo

Examples

set.seed(123)
X = matrix(rnorm(500),50,10)
Y = matrix(rnorm(500),50,10)
X = scale(X, scale = TRUE)
Y = scale(Y, scale = TRUE)
fit <- o2pls(X, Y, 1, 2, 2)
summary(fit)

orthogonal scores algorithn of partial leat squares (opls) projection

Description

orthogonal scores algorithn of partial leat squares (opls) projection

Usage

opls(X, Y, ncomp, maxiter, tol)

Value

a list of opls result

Orthogonal partial least squares discriminant analysis

Description

Computes orthogonal scores partial least squares regressions with the NIPALS algorithm. It return a comprehensive set of pls outputs (e.g. scores and vip).

Usage

oplsda(X, Y, nc, scale = FALSE, center = TRUE, maxiter = 100, tol = 1e-05)

Arguments

X

a O2pls object or a matrix of predictor variables.

Y

a single vector indicate the group

nc

the number of pls components (the one joint components + number of orthogonal components ).

scale

logical indicating whether X must be scaled (suggest TRUE).

center

boolean values determining if data should be centered or not

maxiter

maximum number of iterations.

tol

limit for convergence of the algorithm in the nipals algorithm.

Value

a list containing the following elements:

nc the number of components used(one joint components + number of orthogonal components
scores a matrix of scores corresponding to the observations in X, The components retrieved correspond to the ones optimized or specified.
Xloadings a matrix of loadings corresponding to the explanatory variables. The components retrieved correspond to the ones optimized or specified.
Yloadings a matrix of partial least squares loadings corresponding to Y
vip the VIP matrix.
xvar a matrix indicating the standard deviation of each component (sd), the variance explained by each single component (explained_var) and the cumulative explained variance (cumulative_explained_var). These values are computed based on the data used to create the projection matrices.
projection_matrix the matrix of projection matrix
weight a matrix of partial least squares ("pls") weights.

Author(s)

Kai Guo

Examples

X <- matrix(rnorm(50),10,5)
Y <- matrix(rnorm(50),10,5)
fit <- o2pls(X,Y,2,1,1)
yy <- rep(c(0,1),5)
fit0 <- oplsda(fit,yy,2)

order a vector

Description

order a vector

Usage

order_cpp(x)

Value

a integer vector

order a vector of sting

Description

order a vector of sting

Usage

order_str(x)

Value

An character vector

Score or loading plot for the O2PLS results

Description

Score or loading plot for the O2PLS results

Usage

## S3 method for class 'O2pls'
plot(
  x,
  type = "score",
  var = "Xjoint",
  group = NULL,
  ind = c(1, 2),
  color = NULL,
  top = 20,
  ellipse = TRUE,
  order = FALSE,
  pt.size = 3,
  label = TRUE,
  label.size = 4,
  repel = TRUE,
  rotation = FALSE,
  ...
)

Arguments

x

an O2pls object

type

score or loading

var

specify Xjoint

group

color used for score plot

ind

which components to be used for score plot or loading plot

color

color used for score or loading plot

top

the number of largest loading value to plot

ellipse

TRUE/FALSE

order

order by the value or not

pt.size

point size

label

plot label or not (TRUE/FALSE)

label.size

label size

repel

use ggrepel to show the label or not

rotation

flip the figure or not (TRUE/FALSE)

...

For consistency

Value

a ggplot2 object

Author(s)

Kai Guo

Examples

X <- matrix(rnorm(50),10,5)
Y <- matrix(rnorm(50),10,5)
fit <- o2pls(X,Y,2,1,1)
plot(fit, type="score")

Score, VIP or loading plot for the O2PLS results

Description

Score, VIP or loading plot for the O2PLS results

Usage

## S3 method for class 'o2plsda'
plot(
  x,
  type = "score",
  group = NULL,
  ind = c(1, 2),
  color = NULL,
  top = 20,
  ellipse = TRUE,
  order = FALSE,
  pt.size = 3,
  label = TRUE,
  label.size = 4,
  repel = FALSE,
  rotation = FALSE,
  ...
)

Arguments

x

an o2plsda object

type

score, vip or loading

group

color used for score plot

ind

which components to be used for score plot or loading plot

color

color used for score or loading plot

top

the number of largest loading value to plot

ellipse

TRUE/FALSE

order

order by the value or not

pt.size

point size

label

plot label or not (TRUE/FALSE)

label.size

label size

repel

use ggrepel to show the label or not

rotation

flip the figure or not (TRUE/FALSE)

...

For consistency

Value

a ggplot2 object

Author(s)

Kai Guo

Examples

X <- matrix(rnorm(50),10,5)
Y <- matrix(rnorm(50),10,5)
fit <- o2pls(X,Y,2,1,1)
yy <- rep(c(0,1),5)
fit0 <- oplsda(fit,yy,2)
plot(fit0, type="score", group = factor(yy))

Score, VIP or loading plot for the plsda results

Description

Score, VIP or loading plot for the plsda results

Usage

## S3 method for class 'plsda'
plot(
  x,
  type = "score",
  group = NULL,
  ind = c(1, 2),
  color = NULL,
  top = 20,
  ellipse = TRUE,
  order = FALSE,
  pt.size = 3,
  label = TRUE,
  label.size = 4,
  repel = FALSE,
  rotation = FALSE,
  ...
)

Arguments

x

an plsda object

type

score, vip or loading

group

color used for score plot

ind

which components to be used for score plot or loading plot

color

color used for score or loading plot

top

the number of largest loading value to plot

ellipse

TRUE/FALSE

order

order by the value or not

pt.size

point size

label

plot label or not (TRUE/FALSE)

label.size

label size

repel

use ggrepel to show the label or not

rotation

flip the figure or not (TRUE/FALSE)

...

For consistency

Value

a ggplot2 object

Author(s)

Kai Guo

Examples

X <- matrix(rnorm(500),10,50)
Y <- rep(c("a","b"),each=5)
fit0 <- plsda(X,Y,2)
plot(fit0, type = "score", group = factor(Y))

Partial least squares discriminant analysis

Description

Perform a PLS discriminant analysis

Usage

plsda(X, Y, nc, scale = TRUE, center = TRUE, cv = TRUE, nr_folds = 5)

Arguments

X

a matrix of predictor variables.

Y

a single vector indicate the group

nc

the number of pls components (the one joint components + number of orthogonal components ).

scale

logical indicating whether X must be scaled (suggest TRUE).

center

logical indicating whether X must be centered (suggest TRUE).

cv

logical indicating whether cross-validation will be performed or not (suggest TRUE).

nr_folds

nr_folds Integer to indicate the folds for cross validation.

Value

a list containing the following elements:

nc the number of components used(one joint components + number of orthogonal components
scores a matrix of scores corresponding to the observations in X, The components retrieved correspond to the ones optimized or specified.
Xloadings a matrix of loadings corresponding to the explanatory variables. The components retrieved correspond to the ones optimized or specified.
vip the VIP matrix.
xvar variance explained of X by each single component.
R2Y variance explained of Y by each single component.
PRESS The residual sum of squares for the samples which were not used to fit the model
Q2 quality of cross-validation

Author(s)

Kai Guo

Examples

X <- matrix(rnorm(500),10,50)
Y <- rep(c("a","b"),each=5)
fit <- plsda(X,Y,2)

Print the summary of O2PLS results.

Description

Print the summary of O2PLS results.

Usage

## S3 method for class 'O2pls'
print(x, ...)

Arguments

x

An O2pls object

...

For consistency

Author(s)

Kai Guo

Examples

X <- matrix(rnorm(50),10,5)
Y <- matrix(rnorm(50),10,5)
object <- o2pls(X,Y,1,1,1)
print(object)

Print the summary of plsda results.

Description

Print the summary of plsda results.

Usage

## S3 method for class 'plsda'
print(x, ...)

Arguments

x

An plsda object

...

For consistency

Author(s)

Kai Guo

Examples

X <- matrix(rnorm(500),10,50)
Y <- rep(c("a","b"),each=5)
fit <- plsda(X,Y,2)
print(fit)

calculate RMSE

Description

calculate RMSE

Usage

rcpp_rmse(y, y_hat)

Value

root-mean-square error value

sum square of a matrix

Description

sum square of a matrix

Usage

s2(x)

Value

sum square value of the vector

sampling a vector

Description

sampling a vector

Usage

sample_cpp(x, n)

Value

a vector of length ‘size’ with element drawn from ‘x’

lapply sampling

Description

lapply sampling

Usage

sample_lapply(X, n)

Value

a list

Extract the scores from an O2PLS fit

Description

This function extracts score matrices from an O2PLS fit

Usage

scores(x, ...)

Arguments

x

Object of class O2pls

...

For consistency

Value

Scores matrix

Extract the scores from an O2PLS fit

Description

This function extracts scores parameters from an O2PLS fit

Usage

## S3 method for class 'O2pls'
scores(x, score = c("Xjoint", "Yjoint", "Xorth", "Yorth"), ...)

Arguments

x

Object of class O2pls

score

the scores matrix for one of "Xjoint", "Yjoint", "Xorth", "Yorth"

...

Other arguments

Value

score matrix

Extract the scores from an O2PLS DA analysis

Description

Extract the scores from an O2PLS DA analysis

Usage

## S3 method for class 'o2plsda'
scores(x, ...)

Arguments

x

Object of class o2plsda

...

Other arguments

Value

score matrix

Author(s)

Kai Guo

Extract the scores PLSDA analysis

Description

Extract the scores PLSDA analysis

Usage

## S3 method for class 'plsda'
scores(x, ...)

Arguments

x

Object of class plsda

...

Other arguments

Value

score matrix

Author(s)

Kai Guo

sort string

Description

sort string

Usage

sort_str(strings)

Value

A vector of string

split a vector

Description

split a vector

Usage

split_str(x)

Value

a list of vectors containing the values for the groups

Summary of an O2PLS object

Description

Summary of an O2PLS object

Usage

## S3 method for class 'O2pls'
summary(object, ...)

Arguments

object

a O2pls object

...

For consistency

Value

Detail of O2PLS results

Author(s)

Kai Guo

Examples

X <- matrix(rnorm(50),10,5)
Y <- matrix(rnorm(50),10,5)
object <- o2pls(X,Y,1,1,1)
summary(object)

Summary of an plsda object

Description

Summary of an plsda object

Usage

## S3 method for class 'plsda'
summary(object, ...)

Arguments

object

a plsda object

...

For consistency

Value

Detail of plsda results

Author(s)

Kai Guo

Examples

X <- matrix(rnorm(500),10,50)
Y <- rep(c("a","b"),each=5)
fit <- plsda(X,Y,2)
summary(fit)

unlist a list

Description

unlist a list

Usage

unlist_cpp(list)

Value

a vector of an appropriate mode to hold the list components.

Extract the VIP values from the O2PLS-DA object

Description

Extract the VIP values from the O2PLS-DA object

Usage

vip(x)

Arguments

x

the o2plsda object or plsda object

Value

a data frame