Inferring clonal evolution of tumors from single nucleotide somatic mutations (1210.3384v4)

Abstract: High-throughput sequencing allows the detection and quantification of frequencies of somatic single nucleotide variants (SNV) in heterogeneous tumor cell populations. In some cases, the evolutionary history and population frequency of the subclonal lineages of tumor cells present in the sample can be reconstructed from these SNV frequency measurements. However, automated methods to do this reconstruction are not available and the conditions under which reconstruction is possible have not been described. We describe the conditions under which the evolutionary history can be uniquely reconstructed from SNV frequencies from single or multiple samples from the tumor population and we introduce a new statistical model, PhyloSub, that infers the phylogeny and genotype of the major subclonal lineages represented in the population of cancer cells. It uses a Bayesian nonparametric prior over trees that groups SNVs into major subclonal lineages and automatically estimates the number of lineages and their ancestry. We sample from the joint posterior distribution over trees to identify evolutionary histories and cell population frequencies that have the highest probability of generating the observed SNV frequency data. When multiple phylogenies are consistent with a given set of SNV frequencies, PhyloSub represents the uncertainty in the tumor phylogeny using a partial order plot. Experiments on a simulated dataset and two real datasets comprising tumor samples from acute myeloid leukemia and chronic lymphocytic leukemia patients demonstrate that PhyloSub can infer both linear (or chain) and branching lineages and its inferences are in good agreement with ground truth, where it is available.

• The paper introduces PhyloSub, a novel Bayesian computational method that reconstructs tumor clonal evolution from somatic single nucleotide mutation data.
• The paper also introduces a partial order plot visualization tool to represent phylogenetic uncertainty and demonstrates the method's efficacy on real cancer datasets.
• Results demonstrate PhyloSub's ability to infer clonal lineages aligning with ground truth, holding potential for improved cancer diagnosis and therapy.

Summary of "Inferring clonal evolution of tumors from single nucleotide somatic mutations"

The paper "Inferring clonal evolution of tumors from single nucleotide somatic mutations" introduces a novel computational approach, PhyloSub, designed to reconstruct clonal evolutionary histories of tumors from somatic single nucleotide variant (SNV) frequency data. This approach is necessitated by advances in high-throughput sequencing which can detect SNVs within heterogeneous tumor populations but lacks methods for accurately reconstructing the subclonal histories reflected by these data.

PhyloSub leverages a Bayesian nonparametric model based on a tree-structured stick-breaking process to infer the phylogenetic relationships among subclonal lineages in cancers. The model uniquely incorporates a Dirichlet process prior over trees, allowing it to estimate both the number and the hierarchical relationships of cancer cell subclones directly from observed SNV data. In doing so, PhyloSub can capture both linear and branching evolutionary patterns that reflect the subclonal architecture within tumors.

A significant contribution of this methodological development includes the introduction of a visualization tool, the "partial order plot," which enables representation of phylogenetic uncertainty when multiple evolutionary scenarios are possible given the available SNV frequency data. This provides a nuanced depiction of the probabilistic lineage landscape, allowing researchers to visually assess competing hypotheses supported by the dataset.

The efficacy of PhyloSub is demonstrated through its application to both simulated datasets and real-world cancer datasets—specifically tumors from acute myeloid leukemia (AML) and chronic lymphocytic leukemia (CLL). In these experiments, the model successfully infers clonal lineages that align well with established ground truth obtained through single-cell sequencing assays or previously conducted expert-driven analyses. The results suggest that PhyloSub not only accurately reconstructs phylogenies but also manages uncertainties inherent in the sequencing data effectively.

This work advances the theoretical understanding of clonal evolution in cancer by formalizing conditions under which unique phylogenetic reconstruction is possible using SNVs. The model's assumption of the infinite sites model, which posits that each mutation appears only once and does not revert, plays a pivotal role in constraining the potential phylogenetic solutions and thereby facilitates the automatic inference of robust candidate phylogenies.

Overall, the implications of this research are significant for both theoretical and practical pursuits in oncology. Theoretically, it progresses the framework for understanding the evolutionary dynamics of cancer, contributing to the broader field of phylogenetics as applied to somatic mutation data. Practically, PhyloSub's ability to elucidate subclonal architectures holds potential for influencing diagnostic and therapeutic strategies, where understanding the evolution and heterogeneity of tumors can have direct clinical applications.

Considering possible future developments, the integration and combination of PhyloSub with other genomic data types—such as copy number alterations or epigenomic modifications—could provide enriched evolutionary reconstructions fostering deeper insights into tumor heterogeneity and evolution. Additionally, as high-throughput technologies evolve, the scalability of this approach to larger datasets with more complex mutation landscapes presents a promising avenue for research.