Skip to contents

Wrapper function for complete DRS processing by ninetails package (legacy mode).

Source: R/ninetails_check_tails_guppy.R
check_tails_guppy.Rd

Orchestrates the full ninetails analysis pipeline for direct RNA sequencing (DRS) data basecalled with Guppy and characterised by nanopolish (or tailfindr). The function validates inputs, extracts poly(A) tail features from multi-Fast5 files, segments tails into chunks, computes Gramian Angular Fields (GAFs), classifies reads with a pretrained CNN, and writes output files.

Usage

check_tails_guppy(
  polya_data,
  sequencing_summary,
  workspace,
  num_cores = 1,
  basecall_group = "Basecall_1D_000",
  pass_only = TRUE,
  qc = TRUE,
  save_dir,
  prefix = "",
  part_size = 1000000
)

Arguments

polya_data

Character string or data frame. Full path of the .tsv file produced by nanopolish polya or tailfindr (DRS only), or an in-memory data frame. Tailfindr input is converted automatically.

sequencing_summary

Character string or data frame. Full path of the .txt sequencing summary file, or an in-memory data frame.

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 data should be extracted.

pass_only

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

qc

Logical [TRUE]. If TRUE, terminal non-A residue positions likely arising from nanopolish segmentation errors are labelled with a "-WARN" suffix. It is up to the user whether to include or discard such reads.

save_dir

Character string. Full path of the directory where output files should be stored.

prefix

Character string (optional, default ""). If provided, inserted into output file names between the timestamp and the file-type suffix.

part_size

Numeric [1000000]. Maximum number of rows processed at once. Must be >= 1000. If the input exceeds this value, it is split and processed sequentially.

Value

A named list with two data frames:

read_classes

Data frame. Per-read classification including columns class and comments.

nonadenosine_residues

Data frame. Detailed positional information for all detected non-A residues.

A log file and TSV output files (read_classes, nonadenosine_residues) are also written to save_dir.

Details

Legacy mode. Due to Oxford Nanopore Technologies' transition from Guppy to Dorado, from Fast5 to POD5, and from R9 to R10 chemistry, this pipeline is maintained for backward compatibility only and will not be further optimised. For current data, use check_tails_dorado_DRS instead.

The function accepts either nanopolish or tailfindr output (DRS only). Tailfindr input is automatically converted to nanopolish-like format via check_polya_length_filetype.

For large input files (exceeding part_size rows), the poly(A) table is split into parts and processed sequentially via process_polya_parts to avoid memory overflow.

The pipeline steps are:

  1. Input validation and Fast5 format checking.

  2. Feature extraction (create_tail_feature_list).

  3. Tail segmentation (create_tail_chunk_list).

  4. GAF computation (create_gaf_list).

  5. CNN classification (predict_gaf_classes).

  6. Output creation (create_outputs).

The comments column in the output encodes detailed read classification:

IRL

Insufficient read length.

QCF

Nanopolish QC failed.

MAU

Move transition absent, non-A residue undetected.

MPU

Move transition present, non-A residue undetected.

NIN

Not included in analysis (pass_only = TRUE).

YAY

Move transition present, non-A residue detected.

Filtering criteria: basecaller move value = 1, qc_tag %in% c("PASS") (or c("PASS", "SUFFCLIP")), and nanopolish-estimated tail length >= 10 nt.

See also

check_tails_dorado_DRS for the current (Dorado-based) pipeline, process_polya_complete and process_polya_parts for the internal processing functions, check_polya_length_filetype for input format detection, create_outputs for the output assembly step.

Examples

if (FALSE) { # \dontrun{

results <- ninetails::check_tails_guppy(
  polya_data = system.file('extdata', 'test_data',
                           'nanopolish_output.tsv',
                           package = 'ninetails'),
  sequencing_summary = system.file('extdata', 'test_data',
                                   'sequencing_summary.txt',
                                   package = 'ninetails'),
  workspace = system.file('extdata', 'test_data',
                          'basecalled_fast5',
                          package = 'ninetails'),
  num_cores = 2,
  basecall_group = 'Basecall_1D_000',
  pass_only = TRUE,
  qc = TRUE,
  save_dir = '~/Downloads',
  prefix = "prefix",
  part_size = 2000)

} # }