Detection of somatic structural variants from short-read next-generation sequencing data.

Area of Science:

• Genomics
• Cancer Biology
• Bioinformatics

Background:

• Somatic structural variants (SVs) are crucial in cancer development but difficult to detect using short-read next-generation sequencing (NGS) data.
• Challenges include tumor purity, heterogeneity, limited read length, and alignment ambiguities.
• Despite numerous SV detection tools (callers), each has unique strengths and weaknesses.

Purpose of the Study:

• To review key factors influencing somatic SV detection from NGS data.
• To compare the performance of seven commonly used SV callers.
• To guide the selection of appropriate SV callers for cancer genomics research.

Main Methods:

• Evaluation of seven popular somatic SV callers.
• Analysis of SV detection sensitivity and precision across different SV types and sizes.
• Assessment of performance based on variant allele frequencies and sequencing depth.

Main Results:

• SV caller performance varies significantly depending on SV type, size, variant allele frequency, and sequencing depth.
• Specific reasons for differential performance across various settings were identified.
• Findings provide insights applicable beyond the evaluated callers.

Conclusions:

• Choosing the right SV caller is critical for accurate somatic SV detection in cancer genomics.
• Understanding the factors affecting SV detection performance is essential for robust analysis.
• This review offers a timely guide for researchers in the field.