This function provides a unified interface for all multiblock HD-COX cross-validation methods in the package.
cv.mb.coxmos(
method = c("sb.splsicox", "sb.splsdrcox", "sb.splsdrcox_penalty", "sb.splsdacox",
"isb.splsicox", "isb.splsdrcox", "isb.splsdrcox_penalty", "isb.splsdacox",
"mb.splsdrcox", "mb.splsdacox"),
X,
Y,
max.ncomp = 8,
penalty.list = seq(0.1, 0.9, 0.2),
vector = NULL,
MIN_NVAR = 1,
MAX_NVAR = NULL,
n.cut_points = 5,
n_run = 3,
k_folds = 10,
x.center = TRUE,
x.scale = FALSE,
remove_near_zero_variance = TRUE,
remove_zero_variance = TRUE,
toKeep.zv = NULL,
remove_variance_at_fold_level = FALSE,
remove_non_significant_models = FALSE,
remove_non_significant = FALSE,
alpha = 0.05,
w_AIC = 0,
w_C.Index = 0,
w_AUC = 1,
w_I.BRIER = 0,
MIN_AUC_INCREASE = 0.01,
EVAL_METHOD = "AUC",
pred.method = "cenROC",
pred.attr = "mean",
MIN_AUC = 0.8,
MIN_COMP_TO_CHECK = 3,
max.iter = 200,
times = NULL,
max_time_points = 15,
design = NULL,
fast_mode = FALSE,
MIN_EPV = 5,
return_models = FALSE,
returnData = FALSE,
PARALLEL = FALSE,
n_cores = NULL,
verbose = FALSE,
seed = 123
)A cross-validation object of the specified multiblock type.
Cross-validation method to use: "sb.splsicox", "sb.splsdrcox", "sb.splsdrcox_penalty", "sb.splsdacox", "isb.splsicox", "isb.splsdrcox", "isb.splsdrcox_penalty", "isb.splsdacox", "mb.splsdrcox", or "mb.splsdacox".
List of numeric matrices or data.frames. Explanatory variables. Qualitative variables must be transform into binary variables.
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.
Numeric. Maximum number of PLS components to compute for the cross validation (default: 8).
Numeric vector. Penalty for variable selection for the individual cox models. Variables with a lower P-Value than 1 - "penalty" in the individual cox analysis will be keep for the sPLS-ICOX approach (default: seq(0.1,0.9,0.2)) (sb.splsicox, sb.splsdrcox_penalty, isb.splsicox, isb.splsdrcox_penalty).
Numeric vector. Used for computing best number of variables. As many values as components have to be provided. If vector = NULL, an automatic detection is perform (default: NULL) (sb.splsdrcox, sb.splsdacox, isb.splsdrcox, isb.splsdacox, mb.splsdrcox, mb.splsdacox).
Numeric. Minimum range size for computing cut points to select the best number of variables to use (default: 1) (sb.splsdrcox, sb.splsdacox, isb.splsdrcox, isb.splsdacox, mb.splsdrcox, mb.splsdacox).
Numeric. Maximum range size for computing cut points to select the best number of variables to use (default: NULL) (sb.splsdrcox, sb.splsdacox, isb.splsdrcox, isb.splsdacox, mb.splsdrcox, mb.splsdacox).
Numeric. Number of cut points for searching the optimal number of variables. If only two cut points are selected, minimum and maximum size are used (default: 5) (sb.splsdrcox, sb.splsdacox, isb.splsdrcox, isb.splsdacox, mb.splsdrcox, mb.splsdacox).
Numeric. Number of runs for cross validation (default: 3).
Numeric. Number of folds for cross validation (default: 10).
Logical. If x.center = TRUE, X matrix is centered to zero means (default: TRUE).
Logical. If x.scale = TRUE, X matrix is scaled to unit variances (default: FALSE).
Logical. If remove_near_zero_variance = TRUE, near zero variance variables will be removed (default: TRUE).
Logical. If remove_zero_variance = TRUE, zero variance variables will be removed (default: TRUE).
Character vector. Name of variables in X to not be deleted by (near) zero variance filtering (default: NULL).
Logical. If remove_variance_at_fold_level = TRUE, (near) zero variance will be removed at fold level (default: FALSE).
Logical. If remove_non_significant_models = TRUE, non-significant models are removed before computing the evaluation (default: FALSE).
Logical. If remove_non_significant = TRUE, non-significant variables/components in final cox model will be removed (default: FALSE).
Numeric. Numerical values are regarded as significant if they fall below the threshold (default: 0.05).
Numeric. Weight for AIC evaluator (default: 0).
Numeric. Weight for C-Index evaluator (default: 0).
Numeric. Weight for AUC evaluator (default: 1).
Numeric. Weight for BRIER SCORE evaluator (default: 0).
Numeric. Minimum improvement between different cross validation models to continue evaluating higher values (default: 0.01).
Character. The selected metric will be use to compute the best number of variables (default: "AUC") (sb.splsdrcox, sb.splsdacox, isb.splsdrcox, isb.splsdacox, mb.splsdrcox, mb.splsdacox).
Character. AUC evaluation algorithm method (default: "cenROC") (sb.splsdrcox, sb.splsdacox, isb.splsdrcox, isb.splsdacox, mb.splsdrcox, mb.splsdacox).
Character. Way to evaluate the metric selected (default: "mean").
Numeric. Minimum AUC desire to reach cross-validation models (default: 0.8).
Numeric. Number of penalties/components to evaluate to check if the AUC improves (default: 3).
Numeric. Maximum number of iterations for PLS convergence (default: 200) (sb.splsdrcox, sb.splsdacox, isb.splsdrcox, isb.splsdacox, mb.splsdrcox, mb.splsdacox).
Numeric vector. Time points where the AUC will be evaluated (default: NULL) (sb.splsdrcox, sb.splsdacox, isb.splsdrcox, isb.splsdacox, mb.splsdrcox, mb.splsdacox).
Numeric. Maximum number of time points to use for evaluating the model (default: 15) (sb.splsdrcox, sb.splsdacox, isb.splsdrcox, isb.splsdacox, mb.splsdrcox, mb.splsdacox).
Numeric matrix. Matrix of size (number of blocks in X) x (number of blocks in X) with values between 0 and 1 (default: NULL) (mb.splsdrcox and mb.splsdacox).
Logical. If fast_mode = TRUE, for each run, only one fold is evaluated simultaneously (default: FALSE).
Numeric. Minimum number of Events Per Variable (EPV) you want reach for the final cox model (default: 5).
Logical. Return all models computed in cross validation (default: FALSE).
Logical. Return original and normalized X and Y matrices (default: TRUE).
Logical. Run the cross validation with multicore option (default: FALSE).
Numeric. Number of cores to use for parallel processing. This parameter is only
used if PARALLEL is TRUE. If NULL, it will use all available cores minus one. Otherwise,
it will use the minimum between the value specified and the total number of cores - 1. The fewer
cores used, the less RAM memory will be used.(default: NULL).
Logical. If verbose = TRUE, extra messages could be displayed (default: FALSE).
Number. Seed value for performing runs/folds divisions (default: 123).
cv.sb.splsicox for Single-Block SPLS-ICOX cross-validation,
cv.sb.splsdrcox_penalty for Single-Block SPLS-DRCOX with penalty cross-validation,
cv.sb.splsdrcox for Single-Block SPLS-DRCOX cross-validation,
cv.sb.splsdacox for Single-Block SPLS-DACOX cross-validation,
cv.isb.splsicox for Integrated Single-Block SPLS-ICOX cross-validation,
cv.isb.splsdrcox_penalty for Integrated Single-Block SPLS-DRCOX with penalty cross-validation,
cv.isb.splsdrcox for Integrated Single-Block SPLS-DRCOX cross-validation,
cv.isb.splsdacox for Integrated Single-Block SPLS-DACOX cross-validation,
cv.mb.splsdrcox for Multi-Block SPLS-DRCOX cross-validation,
cv.mb.splsdacox for Multi-Block SPLS-DACOX cross-validation
data("X_multiomic")
data("Y_multiomic")
set.seed(123)
X_train <- X_multiomic
X_train$mirna <- X_train$mirna[1:20,1:30]
X_train$proteomic <- X_train$proteomic[1:20,1:30]
Y_train <- Y_multiomic[1:20,]
cv_mb <- cv.mb.coxmos(method = "sb.splsicox", X = X_train, Y = Y_train,
max.ncomp = 1, n_run = 1, k_folds = 2)
Run the code above in your browser using DataLab