Benchmarking of sequencing technologies defines optimal strategies for genetic variants detection in a human genome

<h4>Background</h4>Advances in sequencing technologies continue to improve the resolution and completeness with which human genetic variation can be characterized. Short-read sequencing remains widely used due to its high base accuracy, throughput, and cost efficiency; however, its limited ability to resolve repetitive and structurally complex regions has accelerated adoption of long-read sequencing platforms, including those from Pacific Biosciences (PacBio) and Oxford Nanopore Technologies (ONT).<h4>Results</h4>We systematically compared sequencing technologies and variant calling pipelines for small variants and structural variants across diverse genomic contexts and sequencing depths. Short-read sequencing combined with DRAGEN achieved high accuracy for single-nucleotide variants (SNVs) and indels in well-mapped and moderately complex regions but showed reduced sensitivity and completeness for structural variant detection. In contrast, long-read sequencing platforms demonstrated clear advantages in detecting structural variants and resolving small variants in difficult genomic regions, although challenges remain in specific indel-prone sequence contexts. Among long-read pipelines, PacBio Revio with DeepVariant achieved the highest SNV and indel accuracy genome-wide, while ONT R10 with DeepVariant performed particularly well in clinically relevant loci. Structural variant detection was dominated by long-read optimized callers, with SVIM and Sawfish performing best for PacBio, and Sniffles2 and CuteSV2 for ONT, consistently outperforming short-read-based methods across variant classes and sizes. Coverage analyses indicated that long-read sequencing reached accuracy saturation between 20 × and 45 × , whereas short-read sequencing required more than 60 × coverage to approach maximal genome completeness.<h4>Conclusions</h4>These results provide practical guidance for platform and pipeline selection. Long-read sequencing enables more comprehensive detection and resolution of structural variants and variation in complex genomic regions, while short-read sequencing remains a cost-effective and scalable solution for high-throughput genotyping and clinically focused applications.