Produces list of GAFs containing exclusively A-nucleotides for neural network training.

Converts A-only signal chunks into Gramian Angular Field (GAF) matrices in parallel. Each chunk is transformed via combine_gafs and the resulting matrices are collected in a flat named list suitable for direct input to the CNN training pipeline.

Usage

create_gaf_list_A(tail_chunk_list, num_cores)

Arguments

tail_chunk_list

List object produced by create_tail_chunk_list_A.

num_cores

Numeric [1]. Number of physical cores to use. Do not exceed 1 less than the number of cores at your disposal.

Value

A named list of GAF matrices, where each element corresponds to one signal chunk and is named <readname>_<index>. Always assign this returned list to a variable; printing the full list to the console may crash the R session.

Details

This training-set variant is designed to work on signal fragments that contain only A residues (produced by create_tail_chunk_list_A). The naming convention for output elements is <readname>_<chunk_index>.

See also

create_tail_chunk_list_A for the preceding pipeline step, combine_gafs for the GAF transformation, create_gaf_list for the production counterpart, prepare_trainingset for the top-level wrapper.

Examples

if (FALSE) { # \dontrun{

create_gaf_list_A(
  tail_chunk_list = tail_chunk_list,
  num_cores = 10)

} # }