Abstract 7345: Double-strand break inference from mosaic chromosomal alterations

Abstract

Abstract Double-strand breaks (DSBs) can cause alterations and rearrangements in chromosomes leading to large structural mutations. There have been previous efforts to identify DSBs in tumors and utilize these as a measure of homologous repair deficiency and response to therapy. In contrast, DSBs have not been well characterized in non-tumor human tissues. To address this void, we explore the distribution of DSBs in hematopoietic stem cells (HSCs) through the analyses of mosaic chromosomal alteration (mCA) breakpoints. Specifically, this work explores the relationship between the likelihood of DSB occurrences and the chromosomes on which they occur. Using data from the UK Biobank, mCA breakpoints were labeled as either DSBs or nonDSBs by proximity to telomeres and centromeres. mCA calls from the UK Biobank were obtained from Loh et al 2020. The START_MB_RANGE and END_MB_RANGE fields were treated as two separate breakpoint events for each entry in the data. Each of these breakpoint ranges was checked to see if it was close to either of the telomeres or to the centromere. Those falling within one Mb were categorized as mCA breakpoints that were likely not a result of DSBs, but rather missegregation of chromosomes or chromosome arms. Ranges that met either of these conditions were labeled “nonDSB” while ranges that did not were labeled “DSB.” These labeled data were processed in R using the Hmisc package. The 1-alpha binomial probability was calculated for the total number of DSBs versus the total number of breakpoint events. The point estimate for each chromosome was plotted along with the upper and lower confidence intervals. This work investigates the properties of DSBs. It is important to note that the observed putative DSBs are not representative of the frequency of DSBs in HSCs, since the resulting mutations can have different impacts on cellular fitness. This point is highlighted by our finding of increased putative DSBs on chromosome 13, as copy number changes involving the 13q14 locus are well documented drivers of chronic lymphocytic leukemia. In this investigation we observe that the proportion of putative DSBs is not correlated with chromosome length (p = 0.2). This result is likely due to the differential fitness effect of mCAs. However, it may also reflect differences in the likelihood of DSBs across the genome. We are currently working on methods from Machine Learning to further understand if there are genetic features which can predict DSBs which give rise to mCAs in HSCs. Citation Format: Abraham Dutch, Cody Bumgardner, Yasminka Jakubek. Double-strand break inference from mosaic chromosomal alterations [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 7345.