Skip to contents

Cross-Validation for weights selection in Transcriptome-Wide Association Studies (TWAS)

Source: R/twas_weights.R
twas_weights_cv.Rd

Performs cross-validation for TWAS, supporting both univariate and multivariate methods. It can either create folds for cross-validation or use pre-defined sample partitions. For multivariate methods, it applies the method to the entire Y matrix for each fold.

Usage

twas_weights_cv(
  X,
  Y,
  fold = NULL,
  sample_partitions = NULL,
  weight_methods = NULL,
  max_num_variants = NULL,
  variants_to_keep = NULL,
  num_threads = 1,
  ...
)

Arguments

X

A matrix of samples by features, where each row represents a sample and each column a feature.

Y

A matrix (or vector, which will be converted to a matrix) of samples by outcomes, where each row corresponds to a sample.

fold

An optional integer specifying the number of folds for cross-validation. If NULL, 'sample_partitions' must be provided.

sample_partitions

An optional dataframe with predefined sample partitions, containing columns 'Sample' (sample names) and 'Fold' (fold number). If NULL, 'fold' must be provided.

weight_methods

A list of methods and their specific arguments, formatted as list(method1 = method1_args, method2 = method2_args), or alternatively a character vector of method names (eg, c("susie_weights", "enet_weights")) in which case default arguments will be used for all methods. methods in the list can be either univariate (applied to each column of Y) or multivariate (applied to the entire Y matrix).

max_num_variants

An optional integer to set the randomly selected maximum number of variants to use for CV purpose, to save computing time.

variants_to_keep

An optional integer to ensure that the listed variants are kept in the CV when there is a limit on the max_num_variants to use.

num_threads

The number of threads to use for parallel processing. If set to -1, the function uses all available cores. If set to 0 or 1, no parallel processing is performed. If set to 2 or more, parallel processing is enabled with that many threads.

Value

A list with the following components:

  • `sample_partition`: A dataframe showing the sample partitioning used in the cross-validation.

  • `prediction`: A list of matrices with predicted Y values for each method and fold.

  • `metrics`: A matrix with rows representing methods and columns for various metrics:

    • `corr`: Pearson's correlation between predicated and observed values.

    • `adj_rsq`: Adjusted R-squared value (which indicates the proportion of variance explained by the model) that accounts for the number of predictors in the model.

    • `pval`: P-value assessing the significance of the model's predictions.

    • `RMSE`: Root Mean Squared Error, a measure of the model's prediction error.

    • `MAE`: Mean Absolute Error, a measure of the average magnitude of errors in a set of predictions.

  • `time_elapsed`: The time taken to complete the cross-validation process.