Compute the transpose multiplication for the relatedness matrix

Usage

.computeTranspose(
  r2,
  transpose_method = "tcrossprod",
  chunk_size = 1000L,
  verbose = FALSE,
  force_symmetric = FALSE,
  config = NULL,
  checkpoint_files = NULL
)

Arguments

r2: a relatedness matrix
transpose_method: character. The method to use for computing the transpose. Options are "tcrossprod", "crossprod", "star", or "chunked"
chunk_size: numeric. If greater than 1 is Number of rows per chunk when transpose_method = "chunked". Defaults to 1000. If less than or equal to 1, the entire matrix is processed in a single chunk.
verbose: logical. If TRUE, print progress through stages of algorithm
force_symmetric: logical. If TRUE, forces the output matrix to be symmetric using Matrix::forceSymmetric. This can be helpful when using chunked multiplication to ensure the final result is exactly symmetric despite potential numerical issues. Defaults to TRUE.
config: list of configuration parameters, used for checkpointing and verbose output
checkpoint_files: list of checkpoint file paths for saving intermediate results when using chunked multiplication. Should contain at least 'tcrossprod_checkpoint' for saving the chunks and 'tcrossprod_ids' for saving the keep_ids used in the tcrossprod checkpoint.

Details

The algorithms and methodologies used in this function are further discussed and exemplified in the vignette titled "examplePedigreeFunctions". For more advanced scenarios and detailed explanations, consult this vignette.