GrassSV - hybrid method to detect structural variants in high throughput DNA-seq data.

Area of Science:

• Genomics
• Bioinformatics
• Population Genetics

Background:

• Genetic diversity, driven by structural variants (SVs), is essential for population adaptation.
• Current SV detection methods often require multiple tools, increasing complexity and computational cost.
• Short-read sequencing data is prevalent due to its affordability, yet comprehensive SV detection remains challenging.

Purpose of the Study:

• To introduce GrassSV, a novel, computationally efficient method for detecting all major structural variant classes.
• To address the limitations of existing SV detection tools, particularly their specialization and computational demands.
• To provide a practical alternative to multi-tool pipelines for SV analysis using short-read data.

Main Methods:

• GrassSV utilizes a hybrid pattern-matching approach for SV detection.
• It integrates depth-of-coverage analysis with contig-based pattern recognition.
• The method is designed for high sensitivity and precision, minimizing false positives and runtime.

Main Results:

• GrassSV demonstrated high accuracy across all major SV types on human and yeast genomes.
• The method achieved lower computational costs compared to existing SV detection tools.
• Robustness was validated using the Genome in a Bottle dataset and synthetic data.

Conclusions:

• GrassSV offers a computationally efficient and accurate solution for comprehensive structural variant detection from short-read sequencing data.
• It provides a practical alternative to complex multi-tool pipelines.
• The tool enhances the analysis of genetic diversity and population adaptation.