Sweepstake evolution revealed by population-genetic analysis of copy-number alterations in single genomes of breast cancer.

Mamoru Kato,Daichi Narushima,Ryuichi Sugino,Daniel A Vasco,Alexander Krasnitz

doi:10.1098/rsos.171060

Abstract

Single-cell sequencing is a promising technology that can address cancer cell evolution by identifying genetic alterations in individual cells. In a recent study, genome-wide DNA copy numbers of single cells were accurately quantified by single-cell sequencing in breast cancers. Phylogenetic-tree analysis revealed genetically distinct populations, each consisting of homogeneous cells. Bioinformatics methods based on population genetics should be further developed to quantitatively analyse the single-cell sequencing data. We developed a bioinformatics framework that was combined with molecular-evolution theories to analyse copy-number losses. This analysis revealed that most deletions in the breast cancers at the single-cell level were generated by simple stochastic processes. A non-standard type of coalescent theory, the multiple-merger coalescent model, aided by approximate Bayesian computation fit well with the data, allowing us to estimate the population-genetic parameters in addition to false-positive and false-negative rates. The estimated parameters suggest that the cancer cells underwent sweepstake evolution, where only one or very few parental cells produced a descendent cell population. We conclude that breast cancer cells successively substitute in a tumour mass, and the high reproduction of only a portion of cancer cells may confer high adaptability to this cancer.

Highlights

The idea that tumour progression can be viewed as a Darwinian process goes back to the 1970s, when it led to the concept of ‘clonal expansion’ [1]
The short sequenced reads are mapped to the human reference genome and copy-number alterations (CNAs) or point mutations present in single tumour cells are identified by bioinformatics analysis
We developed a population-genetic framework combined with bioinformatics techniques that analyses Singlenucleus sequencing (SNS) CNA data, where cancer cells were treated as individuals of a non-sexually reproducing species

Summary

Introduction

The idea that tumour progression can be viewed as a Darwinian process goes back to the 1970s, when it led to the concept of ‘clonal expansion’ [1]. Tumour cells acquire rare advantageous mutations and undergo rapid population expansion due to selection. This evolutionary process in tumours is strongly supported by recent genomic studies employing nextgeneration sequencing for a tumour mass, i.e. a mixture of tumour cells [2,3,4,5,6,7]. In SNS, single cells are isolated from tumour tissue by flow cytometry or micromanipulation, and short DNA fragments (typically 50–200 bp) derived from a single cell are sequenced using a next-generation sequencer. The short sequenced reads are mapped to the human reference genome and copy-number alterations (CNAs) or point mutations present in single tumour cells are identified by bioinformatics analysis. CNAs are detected based on the rationale that a larger number of reads mapping to a genomic region reflects a higher copy number in the region [8,14]

Methods

Results

Conclusion