Skip to contents

SR-TWAS Ensemble Settings

Source: R/twasWeightsPipeline.R
EnsembleParam.Rd

Whether and how twasWeightsPipeline stacks its per-method weights into an SR-TWAS ensemble.

Usage

EnsembleParam(
  enabled = FALSE,
  r2Threshold = 0.01,
  solver = c("quadprog", "glmnet"),
  alpha = 1
)

Arguments

enabled

Logical. Compute SR-TWAS ensemble weights. Default FALSE.

r2Threshold

Minimum cross-validated \(R^2\) for a method to enter the stack. Default 0.01. Stacking needs at least two methods to clear it.

solver

Stacking solver, "quadprog" (default) or "glmnet".

alpha

Elastic-net mixing parameter, used only when solver = "glmnet". Default 1.

Value

A EnsembleParam object, a MethodParam.

Requires cross-validation

Stacking combines each method's out-of-fold predictions, so it cannot run without cross-validation. enabled = TRUE together with CrossValidationParam(folds < 2) is an error. It used to be neither: the ensemble row was simply absent from the result, with ensemble = TRUE still reading as on.

enabled defaults to FALSE because CrossValidationParam defaults to no folds. To get an ensemble, ask for both.

See also

CrossValidationParam

Examples

EnsembleParam(enabled = TRUE, r2Threshold = 0.05)
#> <EnsembleParam>
#>   enabled                  TRUE
#>   r2Threshold              0.05
#>   solver                   quadprog
#>   alpha                    1