Gustaf: Detecting and correctly classifying SVs in the NGS twilight zone

Area of Science:

• Genomics
• Bioinformatics
• Molecular Biology

Background:

• Structural variations (SVs) in genomes, including complex duplications and translocations, remain challenging to fully resolve.
• Existing SV detection tools often show poor agreement and struggle with accurate classification and sizing of variations.

Purpose of the Study:

• To introduce Gustaf (Generic mUlti-SpliT Alignment Finder), a novel SV detection tool.
• To enhance the detection and classification of diverse SVs, particularly in the challenging size range.

Main Methods:

• Development of Gustaf, a generic multi-split alignment strategy for SV detection.
• Identification of SV breakpoints with base-pair resolution.

Main Results:

• Gustaf effectively detects and classifies deletions, inversions, dispersed duplications, and translocations of ≥ 30 bp.
• The tool demonstrates high accuracy for SVs in the 30-100 bp range (the "NGS twilight zone") and larger SVs (>500 bp).
• Gustaf outperforms existing tools in benchmarks and accurately identifies dispersed duplications and translocations, avoiding misclassification.

Conclusions:

• Gustaf provides a robust solution for detecting and classifying a wide spectrum of structural variations.
• The tool's performance, especially in the critical 30-100 bp range, advances genomic analysis capabilities.