Abstract 4289: Long-read, Assembly-Based Characterization of Rearranged Cancer Karyotypes

Ayse Keskus,Tanveer Ahmad,Ataberk Donmez,Yi Xie,Isabel Rodriguez,Rose Milano,Nicole Rossi,Hong Lou,Laksh Malik,Kimberley Billingsley,Cornelis Blauwendraat,Michael Dean,Mikhail Kolmogorov
DOI: https://doi.org/10.1158/1538-7445.am2023-4289
IF: 11.2
2023-01-01
Cancer Research
Abstract:Abstract Introduction: Recent pan-cancer whole-genome sequencing studies revealed the rich landscape of structural variants (SV), from simpler indels to complex events involving multiple breakpoints and sequence gain/loss. SVs may contribute to tumorigenesis through direct modification of coding sequence or deregulation from copy number alterations, enhancer hijacking, or topological domain modification. A substantial part of the variation in the human genome is not accessible to short reads due to mapping ambiguities. Recent benchmarking studies reported that the best short-read methods only have 30-70% SV sensitivity. Long-read sequencing (such as PacBio or Oxford Nanopore) can overcome the limitations of short reads, however the current methods were not designed for the analysis of rearranged cancer genomes with complex copy number profiles. Methods: We developed BGA (Breakpoint Graph Assembler), a method that combines the ideas from long-read assembly and breakpoint graph frameworks. BGA detects abnormally mapped reads and builds a breakpoint graph that characterizes the structure of derived cancer karyotypes. Complex events with multiple breakpoints form connectivity clusters and are classified based on the subgraph properties. BGA also takes advantage of phased haplotypes and can incorporate multiple related datasets (such as in tumor-normal comparison or multi-site tumor sampling). BGA is freely available at: https://github.com/KolmogorovLab/BGA. Results: We first analyzed three cancer cell lines and corresponding matching normal DNAs (HCC1954, H2009, and COLO829). In each cell line, we identified 8-56 somatic rearrangement clusters involving more than two breakpoints and at least 1kb of sequence. In H2009, we identified a chromoplexy event involving chr13 and chr1, consistent with previous FISH experiments. COLO829 showed the lowest number of somatic rearrangement clusters (n=8), including translocation and inversion events between chr3, chr10, and chr12 within the RARB, BICC1, and TRHDE genes. Homologous recombination deficient HCC1954 has the highest number of complex events including the chr17q arm which hosts ERBB2, chromoplexy between chr8, chr5, and chromothripsis in chr21. In addition, we analyzed three other HPV-infected cell lines (CaSki, SCC152, SNU1000). In each of them, we observed complex clusters of HPV-HPV and HPV-human breakpoints that formed cycles, suggesting extrachromosomal amplification. The HPV fragments had many interactions with chromosomal DNA in CaSki and SC152, but not in mostly episomal SNU1000 cells. We also observed karyotype-scale changes that did not involve HPV sequences, such as the simultaneous exchange of six chromosome arms of chr2, chr7, and chr17 in CaSki. Citation Format: Ayse Keskus, Tanveer Ahmad, Ataberk Donmez, Yi Xie, Isabel Rodriguez, Rose Milano, Nicole Rossi, Hong Lou, Laksh Malik, Kimberley Billingsley, Cornelis Blauwendraat, Michael Dean, Mikhail Kolmogorov. Long-read, assembly-based characterization of rearranged cancer karyotypes. [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 4289.
What problem does this paper attempt to address?