splsdrcox {Coxmos}R Documentation

sPLS-DRCOX

Description

This function performs a sparse partial least squares deviance residual Cox (sPLS-DRCOX) (based on plsRcox R package). The function returns a Coxmos model with the attribute model as "sPLS-DRCOX".

Usage

splsdrcox(
  X,
  Y,
  n.comp = 4,
  penalty = 0.5,
  x.center = TRUE,
  x.scale = FALSE,
  remove_near_zero_variance = TRUE,
  remove_zero_variance = FALSE,
  toKeep.zv = NULL,
  remove_non_significant = FALSE,
  alpha = 0.05,
  MIN_EPV = 5,
  returnData = TRUE,
  verbose = FALSE
)

Arguments

X

Numeric matrix or data.frame. Explanatory variables. Qualitative variables must be transform into binary variables.

Y

Numeric matrix or data.frame. Response variables. Object must have two columns named as "time" and "event". For event column, accepted values are: 0/1 or FALSE/TRUE for censored and event observations.

n.comp

Numeric. Number of latent components to compute for the (s)PLS model (default: 10).

penalty

Numeric. Penalty for sPLS-DRCOX. If penalty = 0 no penalty is applied, when penalty = 1 maximum penalty (no variables are selected) based on 'plsRcox' penalty. Equal or greater than 1 cannot be selected (default: 0.5).

x.center

Logical. If x.center = TRUE, X matrix is centered to zero means (default: TRUE).

x.scale

Logical. If x.scale = TRUE, X matrix is scaled to unit variances (default: FALSE).

remove_near_zero_variance

Logical. If remove_near_zero_variance = TRUE, near zero variance variables will be removed (default: TRUE).

remove_zero_variance

Logical. If remove_zero_variance = TRUE, zero variance variables will be removed (default: TRUE).

toKeep.zv

Character vector. Name of variables in X to not be deleted by (near) zero variance filtering (default: NULL).

remove_non_significant

Logical. If remove_non_significant = TRUE, non-significant variables/components in final cox model will be removed until all variables are significant by forward selection (default: FALSE).

alpha

Numeric. Numerical values are regarded as significant if they fall below the threshold (default: 0.05).

MIN_EPV

Numeric. Minimum number of Events Per Variable (EPV) you want reach for the final cox model. Used to restrict the number of variables/components can be computed in final cox models. If the minimum is not meet, the model cannot be computed (default: 5).

returnData

Logical. Return original and normalized X and Y matrices (default: TRUE).

verbose

Logical. If verbose = TRUE, extra messages could be displayed (default: FALSE).

Details

The sPLS-DRCOX function implements the sparse partial least squares deviance residual Cox (sPLS-DRCOX) model, a specialized approach tailored for survival analysis. This method integrates the strengths of the sparse partial least squares (sPLS) technique with the Cox proportional hazards model, leveraging deviance residuals as a bridge.

The function's core lies in its ability to handle high-dimensional data, often encountered in genomics or other omics studies. By incorporating the penalty parameter, which governs the sparsity level, the function offers a fine-grained control over variable selection. This ensures that only the most informative predictors contribute to the model, enhancing interpretability and reducing overfitting.

Data preprocessing is seamlessly integrated, with options to center and scale the predictors, and to remove variables exhibiting near-zero or zero variance. The function also provides a mechanism to retain specific variables, regardless of their variance, ensuring that domain-specific knowledge can be incorporated.

The output is comprehensive, detailing both the sPLS and Cox model components. It provides insights into the selected variables, their contributions across latent components, and the overall fit of the survival model. This rich output aids in understanding the underlying relationships between predictors and survival outcomes.

The sPLS-DRCOX function is grounded in established methodologies and is a valuable tool for researchers aiming to unravel complex survival associations in high-dimensional datasets.

Value

Instance of class "Coxmos" and model "sPLS-DRCOX". The class contains the following elements: X: List of normalized X data information.

Y: List of normalized Y data information.

survival_model: List of survival model information.

penalty: Penalty value selected.

n.comp: Number of components selected.

var_by_component: Variables selected in each PLS component.

call: call function

X_input: X input matrix

Y_input: Y input matrix

B.hat: sPLS beta matrix

R2: sPLS R2

SCR: sPLS SCR

SCT: sPLS SCT

alpha: alpha value selected

nsv: Variables removed by cox alpha cutoff.

nzv: Variables removed by remove_near_zero_variance or remove_zero_variance.

nz_coeffvar: Variables removed by coefficient variation near zero.

class: Model class.

time: time consumed for running the cox analysis.

Author(s)

Pedro Salguero Garcia. Maintainer: pedsalga@upv.edu.es

References

Bastien P (2008). “Deviance residuals based PLS regression for censored data in high dimensional setting.” Chemometrics and Intelligent Laboratory Systems. doi:10.1016/j.chemolab.2007.09.009, https://www.sciencedirect.com/science/article/abs/pii/S0169743907001931?via%3Dihub. Bastien P, Bastien P, Bertrand F, Meyer N, Meyer N, Meyer N, Maumy-Bertrand M (2015). “Deviance residuals-based sparse PLS and sparse kernel PLS regression for censored data.” Bioinformatics. https://academic.oup.com/bioinformatics/article/31/3/397/2366078.

Examples

data("X_proteomic")
data("Y_proteomic")
X <- X_proteomic[,1:50]
Y <- Y_proteomic
splsdrcox(X, Y, n.comp = 3, penalty = 0.25, x.center = TRUE, x.scale = TRUE)
[Package Coxmos version 1.0.2 Index]