Severus detects somatic structural variation and complex rearrangements in cancer genomes using long-read sequencing
Keskus et al. developed Severus, a breakpoint graph-based algorithm for somatic structural variant (SV) calling from Oxford Nanopore and PacBio data. Severus outperformed other long- and short-read SV callers and identified clinically relevant rearrangements in tumour samples that were missed by standard genomic panels. This research demonstrates the value of accurate SV calling for precision oncology and provides new tools and data to advance cancer genomics in the future.
Key points:
Severus works with tumour-only and matched normal samples, produces haplotype-specific SV calling, characterises complex SV patterns, and supports unbalanced cancer karyotypes
Severus identified clinically relevant, cryptic SVs in paediatric leukaemia and lymphoma research samples that were missed by routine clinical tests
Using a phased breakpoint graph, Severus reconstructed complex genomic events such as chromoplexy
This algorithm outperforms existing SV detection methods, which struggle with the complex rearrangements and heterogeneity associated with tumour SV calling
Watch the talk by Mikhail Kolmogorov and Ayse Keskus at AACR 2024
Sample type: tumour-normal cell lines, blood cancer clinical research samples
