Skip to contents

Extracts features of poly(A) tails containing only A nucleotides for training-set preparation.

Source: R/ninetails_training_dataset_production_functions.R
create_tail_feature_list_A.Rd

Training-set variant of the feature extraction wrapper designed exclusively for A-only signals. It processes all reads in parallel via extract_tail_data_trainingset, then applies an inverse filtering criterion: only reads whose pseudomove vectors do not contain consecutive non-zero runs of length >= 4 are retained. This ensures the resulting dataset represents pure homopolymer A tails without modification artefacts.

Usage

create_tail_feature_list_A(
  nanopolish,
  sequencing_summary,
  workspace,
  num_cores,
  basecall_group,
  pass_only = TRUE
)

Arguments

nanopolish

Character string. Full path of the .tsv file produced by nanopolish polya.

sequencing_summary

Character string. Full path of the .txt file with the sequencing summary.

workspace

Character string. Full path of the directory containing basecalled multi-Fast5 files.

num_cores

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

basecall_group

Character string ["Basecall_1D_000"]. Name of the level in the Fast5 file hierarchy from which the data should be extracted.

pass_only

Logical [TRUE]. If TRUE, only reads tagged by nanopolish as "PASS" are retained. Otherwise, reads tagged as "PASS" or "SUFFCLIP" are included.

Value

A named list with one element:

tail_feature_list

Named list of per-read feature lists (as returned by extract_tail_data_trainingset) containing only reads with pure A tails.

Always assign this returned list to a variable; printing the full list to the console may crash the R session.

Details

The inverse filtering uses a sliding-window approach (stats::embed) to detect runs of >= 4 consecutive non-zero pseudomoves. Reads that pass this filter (i.e. have no such runs) are collected as the A-only reference set. Unlike create_tail_feature_list_trainingset, the returned list does not include zeromoved_readnames or nonpseudomoved_readnames categories.

See also

create_tail_feature_list_trainingset for the non-A variant, create_tail_chunk_list_A for the next pipeline step, extract_tail_data_trainingset for per-read extraction, prepare_trainingset for the top-level wrapper.

Examples

if (FALSE) { # \dontrun{

create_tail_feature_list_A(
  nanopolish = '/path/to/file',
  sequencing_summary = '/path/to/file',
  workspace = '/path/to/guppy/workspace',
  num_cores = 10,
  basecall_group = 'Basecall_1D_000')

} # }