Structural variant detection across sequencing technologies.

Repetitiveness- and signature-density-aware breakpoint recognition followed by haplotype-resolved local assembly

The authors introduce HitSV, a structural variant (SV) detection tool that couples repetitiveness- and signature-density-aware breakpoint recognition with haplotype-resolved local assembly, enabling base-resolution SV reconstruction and genotyping across sequencing technologies. On the HG002 benchmark, its genotype F1 score exceeded that of state-of-the-art SV callers by 15-72% for long reads, 10-36% for short reads and 12% for hybrid sequencing. Applied to the 1000 Genomes Project (1KGP) phase 4 cohort, HitSV expanded the SV callset by 31.5%, substantially reshaping allele-frequency landscapes. In a large Chinese long-read cohort, it uncovered tandem-repeat-mobile-element composite arrays as a prevalent, multi-allelic and previously underrecognized class of complex SVs.

Synthesis written by Geno'X. For the full original abstract, please refer to the source publication.

A single caller that performs well on short-read, long-read and hybrid data simplifies life for laboratories running several technologies side by side. The gain is measured on HG002 and population cohorts: no diagnostic-yield data on patients is presented. The 31.5% expansion of the 1KGP callset must flow into the frequency databases used to filter SVs, otherwise common but previously undetected SVs may be misread as rare.

Analysis by Dr Thibaut Benquey