LongReadSum: A fast and flexible quality control and signal summarization tool for long-read sequencing data

Jonathan Perdomo, Mian Umair Ahsan, Qian Liu, Fang Li, Kai WangbioRxiv 2024 · DOI

TL;DR

LongReadSum is a new computational tool designed to quickly and comprehensively assess the quality of long-read sequencing data from various platforms. It addresses the growing need for efficient analysis of massive datasets, offering detailed reports on read quality, alignment, and even raw signal data, which is crucial for advanced applications like methylation analysis.

What Problem Does This Paper Address?

The field of long-read sequencing (think Oxford Nanopore and PacBio) is exploding, generating vast amounts of data that are essential for applications ranging from whole-genome sequencing to detecting DNA modifications. However, existing quality control (QC) tools, many of which were built for shorter reads, struggle to keep up. They often lack support for the diverse file formats produced by these new technologies, can't handle the sheer scale of data from high-throughput sequencers like the Oxford Nanopore PromethION, and don't offer comprehensive metrics, especially for raw signal data which is key for advanced analyses like base modification detection. This creates a bottleneck, making it difficult for researchers to quickly assess data quality and identify potential issues before diving into complex downstream analyses.

Methods

LongReadSum is a multi-threaded C++ tool designed for speed and flexibility. It processes various long-read data formats, including FASTA, FASTQ, BAM (aligned and unaligned), FAST5, and the newer POD5 format. A key innovation is its ability to analyze raw signal intensity data from Oxford Nanopore, which is crucial for applications like methylation detection. The tool generates a comprehensive HTML report summarizing metrics across different aspects of the sequencing data, such as read length distributions, base quality, alignment statistics, and base modification information. It leverages C++ for core computations and Python for reporting, with a SWIG wrapper for seamless integration.

Datasets

The paper evaluates LongReadSum using publicly available datasets, including:

  • An aligned BAM file from the HG002 genome sequenced on an Oxford Nanopore PromethION flowcell (57 gigabases).
  • A whole-genome sequencing (WGS) dataset of the HG002 genome from an Oxford Nanopore MinION platform.
  • Data from various applications including cDNA sequencing, direct mRNA sequencing, reduced representation methylation sequencing (RRMS), and whole-genome sequencing (WGS) across different platforms (ONT R9.4.1, ONT R10.4.1, and Illumina Complete Long Read).

Key Insights and Ideas

  • The "Big Data" Challenge: Long-read sequencing platforms are generating terabytes of data per flowcell, far exceeding the capacity of many existing QC tools. LongReadSum's design specifically addresses this scalability issue.
  • Beyond Basecalls: The Power of Signal: Raw signal data (from FAST5/POD5 files) contains rich information beyond just the sequenced bases. LongReadSum's ability to analyze this signal data opens doors for advanced applications like identifying DNA modifications (e.g., methylation) directly from the sequencing signal.
  • Platform Agnosticism is Key: Researchers often use multiple sequencing technologies. A single tool that can handle data from PacBio, Oxford Nanopore, and Illumina long-read platforms simplifies the QC workflow significantly.
  • Speed Matters: For routine QC, especially in high-throughput labs, speed is paramount. LongReadSum's multi-threaded C++ implementation delivers substantial performance improvements over existing tools.

Key Results

  • Performance Boost: LongReadSum achieved a significant speedup compared to NanoPlot, a leading long-read QC tool. For a 57-gigabase dataset, LongReadSum completed QC in approximately 15 minutes, while NanoPlot took about 1 hour and 37 minutes on the same hardware.
  • Comprehensive Reporting: The tool provides detailed QC reports covering read statistics, base quality, alignment metrics, and even signal intensity plots for specific regions of interest.
  • Broad Data Format Support: LongReadSum supports a wide array of file formats from major long-read sequencing platforms, including FASTA, FASTQ, BAM, FAST5, and POD5.
  • Demonstrated Use Cases: The paper showcases LongReadSum's utility across diverse applications, including cDNA sequencing, direct mRNA sequencing, RRMS, and WGS, highlighting its flexibility.

Limitations and Future Work

  • Evolving Standards: The authors acknowledge that sequencing file formats are constantly evolving. They plan to continue developing LongReadSum to support new high-performance formats and emerging technologies like optical maps.
  • Deeper Analysis: Future work includes adding more in-depth mapping and base quality QC statistics, particularly for analyzing multiple sequencing runs on the same sample or for multiplexed samples.
  • Platform Expansion: While LongReadSum covers major platforms, future development may include support for additional emerging sequencing technologies.

Summary updated

Abstract

Abstract While several well-established quality control (QC) tools are available for short reads sequencing data, there is a general paucity of computational tools that provide long read metrics in a fast and comprehensive manner across all major sequencing platforms (such as PacBio, Oxford Nanopore, Illumina Complete Long Read) and data formats (such as ONT POD5, FAST5, basecall summary files and PacBio unaligned BAM). Additionally, none of the current tools provide support for summarizing Oxford Nanopore basecall signal or comprehensive base modification (methylation) information from genomic data. Furthermore, nowadays a single PromethION flowcell on the Oxford Nanopore platform can generate terabytes of signal data, which cannot be handled by existing tools designed for small-scale flowcells. To address these challenges, here we present LongReadSum, a multi-threaded C++ tool which provides fast and comprehensive QC reports on all major aspects of sequencing data (such as read, base, base quality, alignment, and base modification metrics) and produce basecalling signal intensity information from the Oxford Nanopore platform. We demonstrate use cases to analyze cDNA sequencing, direct mRNA sequencing, reduced representation methylation sequencing (RRMS) through adaptive sequencing, as well as whole genome sequencing (WGS) data using diverse long-read platforms.
BibTeX
@misc{perdomo2024longreadsum,
  author = {Jonathan Perdomo and Mian Umair Ahsan and Qian Liu and Fang Li and Kai Wang},
  title = {{LongReadSum: A fast and flexible quality control and signal summarization tool for long-read sequencing data}},
  year = {2024},
  doi = {10.1101/2024.08.05.606643},
  howpublished = {bioRxiv},
}
Figure