Long-Read Sequencing Outperforms Short-Read Sequencing in Detecting Most Structural Variations

Area of Science:

• Genomics
• Cancer Biology
• Bioinformatics

Background:

• Structural variations (SVs) are prevalent in cancer genomes and influence tumorigenesis.
• Short-read sequencing (SRS) has been the primary method for SV detection, but faces limitations.
• Long-read sequencing (LRS) technologies offer new possibilities for comprehensive SV analysis.

Purpose of the Study:

• To compare the efficacy of LRS and SRS in detecting structural variations in a prostate cancer cell line.
• To identify the strengths and weaknesses of each sequencing technology for SV detection.
• To evaluate the accuracy and precision of breakpoint identification by LRS and SRS.

Main Methods:

• Whole-genome sequencing (WGS) of the LNCaP prostate cancer cell line using Oxford Nanopore Technology (LRS).
• Comparison of SVs detected by LRS with publicly available LNCaP SRS data.
• Bioinformatic analysis to assess SV detection accuracy, breakpoint precision, and types of variations identified.

Main Results:

• LRS demonstrated superiority in detecting insertions of all sizes and deletions <1000 bp.
• SRS proved effective for identifying long deletions (>1000 bp) due to paired-end reads.
• LRS provided more precise breakpoint identification and fewer false positives for duplications and inversions compared to SRS.

Conclusions:

• LRS generally outperforms SRS in detecting most structural variations, offering higher accuracy and precision.
• LRS is particularly advantageous for resolving complex genomic rearrangements in cancer.
• While SRS remains valuable for specific applications like long deletion detection, LRS represents a significant advancement for comprehensive cancer genome analysis.