Abstract 2326: Integrating single nucleotide variants (SNVs), copy number alterations (CNAs), and structural variants (SVs) into single-cell clonal lineage inference

Abstract Background: We extend algorithms for clonal lineage inference (``tumor phylogenetics'') from single-cell DNA sequence (scDNA-seq) data to accommodate broader classes of genetic variants, including single nucleotide variants (SNVs), copy number alterations (CNAs), and structural variants (SVs). Methods: We develop a computational method based on constrained optimization, posing a problem of inferring cellular lineage trees consistent with a given scDNA-seq data set constrained to follow a Dollo parsimony model for SNVs and SVs. We then seek an optimal tree consistent with these constraints so as to optimize for a CNA minimum evolution model balanced against measure of consistency of the different data types with one another relative to the reconstructed set of clones. We pose this problem as an integer linear program (ILP) that we solve heuristically via a coordinate descent algorithm. We tested the resulting method on a combination of synthetic scDNA-seq data and publicly available real data. Results: Results on synthetic data show the method to be effective at constructing lineage trees under a variety of parameter domains, including variations in noise levels, clone numbers, single cells, and mutation rates. We further show it to yield improved accuracy over widely used prior models that operate on only subsets of these variant types. Application to previously published scDNA-seq data demonstrate the method to be effective at inferring clonal lineage trees incorporating all three variant types. The results further show how the different kinds of variants can contribute in tandem to progression along single cell lineages or to the emergence of clones exhibiting hallmarks of distinct mechanisms of progression. Conclusion: The results support the effectiveness of our method in resolving accurate trees from scDNA-seq data including diverse variant types, as well as the value of considering these mutation types collectively in developing a comprehensive understanding of how various forms of somatic mutability together shape clonal evolution in cancers. Citation Format: Nishat Bristy, Xuecong Fu, Russell Schwartz. Integrating single nucleotide variants (SNVs), copy number alterations (CNAs), and structural variants (SVs) into single-cell clonal lineage inference 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 (6Suppl): Abstract nr 2326.