Skip to contents

Complete Oxford Nanopore poly(A)/poly(T) tail analysis pipeline for Dorado cDNA data.

Source: R/check_tails_dorado_cDNA.R
check_tails_dorado_cDNA.Rd

This function extends the check_tails_dorado_DRS pipeline to handle cDNA sequencing data by adding BAM file processing for sequence extraction and Dorado-style read orientation classification. The pipeline identifies and characterizes non-adenosine nucleotides within poly(A) and poly(T) tails using the same signal processing and machine learning approach as the DRS pipeline, but with automatic classification of read orientations.

Usage

check_tails_dorado_cDNA(
  bam_file,
  dorado_summary,
  pod5_dir,
  num_cores = 1,
  qc = TRUE,
  save_dir,
  prefix = "",
  part_size = 40000,
  cleanup = FALSE
)

Arguments

bam_file

Character string. Path to the BAM file containing aligned cDNA reads with basecalled sequences. This file will be split into parts for memory management.

dorado_summary

Character string or data frame. Path to Dorado summary file or data frame containing per-read summary information. Must include standard columns such as read_id, filename, etc.

pod5_dir

Character string. Path to directory containing POD5 files with raw nanopore signal data corresponding to the reads in the summary file.

num_cores

Integer [1]. Number of CPU cores to use for parallel processing. Recommend using `parallel::detectCores() - 1` for optimal performance while maintaining system responsiveness.

qc

Logical [TRUE]. Whether to apply quality control filtering during analysis. When TRUE, applies standard ninetails QC filters including tail length filtering and coordinate validation.

save_dir

Character string. Path to directory where all output files and intermediate results will be saved. Directory will be created if it doesn't exist.

prefix

Character string [""]. Optional prefix to add to all output file names. Useful for distinguishing between different experimental conditions or samples.

part_size

Integer [40000]. Number of reads to process in each chunk when splitting large input files. Larger values use more memory but may be faster. Adjust based on available system memory.

cleanup

Logical [FALSE]. Whether to remove intermediate files after successful pipeline completion. When FALSE, all intermediate files are preserved for inspection.

Value

A named list containing the final analysis results:

read_classes

Data frame with per-read classification results including readname, contig, poly(A) length, QC tag, class, comments, and tail_type

nonadenosine_residues

Data frame with predicted non-adenosine positions within poly(A)/poly(T) tails including readname, contig, prediction, estimated position, poly(A) length, QC tag, and tail_type

Pipeline Overview

The cDNA pipeline follows the same analysis flow as check_tails_dorado_DRS with these additions:

  1. BAM Processing: Extracts basecalled sequences from BAM file (required for cDNA data)

  2. Dorado-Style Read Classification: Classifies reads as polyA, polyT, or unidentified using edit distance matching

  3. Standard Processing: Processes reads using the same signal processing and analysis as DRS pipeline

  4. Output with Tail Types: Produces standard read_classes and nonadenosine_residues with tail_type information

Input Requirements

This pipeline requires specific input formats:

  • Dorado Summary: Must contain standard columns for read information

  • BAM File: Aligned cDNA reads with basecalled sequences

  • POD5 Files: Raw signal files corresponding to reads in summary

Key Differences from DRS Pipeline

  • BAM Input: Additional BAM file input for sequence extraction since cDNA data requires basecalled sequences

  • Dorado-Style Read Orientation Classification: Uses edit distance matching of SSP/VNP primers to classify reads as polyA, polyT, or unidentified before processing

Examples

if (FALSE) { # \dontrun{
# Basic cDNA analysis
results <- ninetails::check_tails_dorado_cDNA(
  bam_file = "path/to/aligned_cdna.bam",
  dorado_summary = "path/to/dorado_summary.txt",
  pod5_dir = "path/to/pod5_files/",
  num_cores = 4,
  save_dir = "path/to/output/"
)

# Access results
head(results$read_classes)
head(results$nonadenosine_residues)

# Analysis with custom settings
results <- ninetails::check_tails_dorado_cDNA(
  bam_file = "large_dataset.bam",
  dorado_summary = summary_df,  # Can pass data frame
  pod5_dir = "/data/pod5/",
  num_cores = 8,
  qc = TRUE,
  save_dir = "/results/experiment1/",
  prefix = "exp1_sample_A",
  part_size = 20000,  # Smaller chunks for limited memory
  cleanup = TRUE      # Remove intermediate files
)
} # }