COMPASS: joint copy number and mutation phylogeny reconstruction from amplicon single-cell sequencing data

Abstract

Reconstructing the history of somatic DNA alterations can help understand the evolution of a tumor and predict its resistance to treatment. Single-cell DNA sequencing (scDNAseq) can be used to investigate clonal heterogeneity and to inform phylogeny reconstruction. However, most existing phylogenetic methods for scDNAseq data are designed either for single nucleotide variants (SNVs) or for large copy number alterations (CNAs), or are not applicable to targeted sequencing. Here, we develop COMPASS, a computational method for inferring the joint phylogeny of SNVs and CNAs from targeted scDNAseq data. We evaluate COMPASS on simulated data and apply it to several datasets including a cohort of 123 patients with acute myeloid leukemia. COMPASS detected clonal CNAs that could be orthogonally validated with bulk data, in addition to subclonal ones that require single-cell resolution, some of which point toward convergent evolution.

Integrating SNVs and CNAs on a phylogenetic tree from single-cell DNA sequencing data

“…We developed a simulator described in the Supplemental Methods, Section 1 , to comprehensively test SCsnvcna under different conditions. We varied 15 variables in the simulation and broke them into five groups (Methods; Supplemental Methods, Section 2 ; Supplemental Table S1 ) for testing SCsnvcna and compared them with seven state-of-the-art methods: SCARLET ( Satas et al 2020 ), COMPASS ( Sollier et al 2023 ), SiFit ( Zafar et al 2017 ), SCG ( Roth et al 2016 ), RobustClone ( Chen et al 2020 ), SiCloneFit ( Zafar et al 2019 ), and BitSC 2 ( Supplemental Methods, Section 3 ; Chen et al 2022 ). Among these methods, SiFit and SiCloneFit only take SNVs as the input and do not infer the phylogenetic tree based on any CNA signal.…”

“…To further show SCsnvcna's robustness to different data sets, we used a different simulator from ours as reported by Sollier et al (2023) to generate additional simulated data sets for orthogonal validation ( Supplemental Methods, Section 5 ). In total, we simulated three data sets using the simulator reported by Sollier et al (2023) to test the robustness of our method, varying the number of cells, number of mutations, and number of nodes on the tree ( Supplemental Methods, Section 5 ). We then ran SCsnvcna on the three sets of the simulated data in two ways, with and without constraining the SNV cell placement, and compared with COMPASS and BiTSC 2 .…”

“…Because of the read depth nonuniformity on the sequences of SNV cells, the detection of CNAs has to be on a very large scale, leading to a low-resolution CNA tree to start with. BiTSC 2 ( Chen et al 2022 ) and COMPASS ( Sollier et al 2023 ) are the other two methods that jointly infer a phylogenetic tree from both SNVs and CNAs. However, like SCARLET, they also require that CNAs and SNVs are from the same set of cells.…”

oncotree2vec – A method for embedding and clustering of tumor phylogenetic trees

“…In this paper we applied oncotree2vec to mutation trees that contain point mutations. However, our method can be easily expanded to copy number trees and to mutation trees containing any sets of genomic alterations such as mixed trees of joint copy number and point mutations [47], or augmented trees [17]. Learning tree embeddings will facilitate the integration of tumor mutation trees with data from other technologies in downstream analysis (e.g., in deep learning approaches for single-cell multi-omics data integration).…”

“…Tumor mutation trees are rooted trees where the nodes represent genomic events such as point mutations, copy number alterations (CNA), or other sets of genomic alterations, and the nodes are connected according to their order in the evolutionary history. Mutation trees can be obtained from single-cell data using several tools, such as SCITE [23], OncoNEM [43] and and HUNTRESS [28] for point mutation trees, SCICoNE [27] and CONET [33] for copy number trees, or COMPASS for joint copy number and point mutation trees [47].…”