Classify multiple cDNA read orientations using Dorado-style poly tail detection

This function analyzes Nanopore cDNA sequences using the Dorado-style approach for poly tail type detection. It uses edit distance-based matching of SSP and VNP primer sequences at both ends of reads, testing forward and reverse orientations to determine if reads contain poly(A) or poly(T) tails. The method uses sliding window matching and requires both score and separation thresholds to be met. The tail_type column is added to original sequence files and all sequences are returned as a single tibble.

Usage

detect_orientation_multiple(sequence_files, num_cores = 1, cli_log = message)

Arguments

sequence_files

Character vector. Paths to sequence files (TSV format) containing read_id and sequence columns.

num_cores

Integer. Number of CPU cores to use for parallel processing.

cli_log

Function for logging messages and progress.

Value

Data frame/tibble containing all sequences with added tail_type column ("polyA", "polyT", or "unidentified") based on Dorado-style edit distance matching of SSP and VNP primers with score and separation validation. The tail_type column is also written back to the original sequence files for persistence.

Examples

if (FALSE) { # \dontrun{
classified_sequences <- detect_orientation_multiple(
  sequence_files = c("sequences_part1.tsv", "sequences_part2.tsv"),
  num_cores = 4,
  cli_log = message
)
} # }