Reconstruction of chromosome-scale copy number profiles of tumor genomes with long-read sequencing | LC26

Abstract
Cancer genomes are characterized by complex karyotypes harboring somatic structural variants (SVs) and copy number alterations (CNAs). While long-read sequencing offers key advantages over short reads — including improved mappability, phasing range, and resolution of complex rearrangements — dedicated computational tools for cancer-specific analysis have been lacking. We present two complementary methods developed for long-read cancer genome analysis. Severus is a breakpoint graph-based somatic SV caller that supports unbalanced karyotypes, characterizes complex multibreak rearrangements, and produces haplotype-specific calls, outperforming existing methods in F1 score on a multitechnology cell line panel. Wakhan enables haplotype-specific CNA calling, reconstructing chromosome-scale profiles of rearranged cancer genomes with superior consistency compared to short- and long-read alternatives. Together, Severus and Wakhan provide an integrated framework for comprehensive long-read cancer genome characterization, with demonstrated applications in breast cancer and pediatric malignancies.

Biography
Mikhail Kolmogorov is currently a tenure-track Stadtman Investigator at the National Cancer Institute, where he leads a group focusing on computational and cancer genomics. Prior to that Mikhail was a postdoctoral fellow at the University of California Santa Cruz, supervised by Dr Benedict Paten. He completed his PhD in September 2019 in Computer Science from University California San Diego, under the mentorship of Dr Pavel Pevzner. Mikhail received his M.S. in bioinformatics from St. Petersburg Academic University, Russia.