Genotyping inversions and tandem duplications.

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Next Generation Sequencing (NGS) facilitates the study of structural genomic variants (SVs), including duplications and inversions, in large populations.
• SVs are implicated in various diseases, notably cancer.
• The increasing availability of NGS data and variant databases highlights the need for robust SV genotyping tools, a need currently unmet.

Purpose of the Study:

• To introduce DIGTYPER, a novel statistical approach for genotyping structural genomic variants (SVs).
• To enable accurate genotyping of duplications and inversions using Next Generation Sequencing (NGS) data.
• To provide a tool that can genotype known SVs from databases without requiring SV re-discovery.

Main Methods:

• DIGTYPER employs a statistical approach to compute genotype likelihoods for given inversions or duplications.
• It integrates breakpoint-spanning read pairs and split alignments for genotyping, accommodating small SV events.
• The method was validated using simulated and real sequencing data.

Main Results:

• DIGTYPER demonstrates favorable performance, particularly for duplications of all sizes and inversions up to 300 bp.
• It was compared against DELLY and SVTyper, showing competitive or superior results.
• DIGTYPER can genotype SVs directly from existing databases, unlike DELLY which requires SV re-discovery.

Conclusions:

• DIGTYPER offers a significant advancement in the accurate and efficient genotyping of structural genomic variants from NGS data.
• The tool addresses the current gap in SV genotyping capabilities, facilitating large-scale genomic studies.
• Its ability to genotype known variants streamlines analysis pipelines and enhances the utility of variant databases.