Integrative inference of subclonal tumour evolution from single-cell and bulk sequencing data

Salem Malikic,S Cenk Sahinalp,Jack Kuipers,Niko Beerenwinkel,Katharina Jahn

doi:10.1038/s41467-019-10737-5

Salem Malikic, S Cenk Sahinalp + Show 3 more

Open Access

https://doi.org/10.1038/s41467-019-10737-5

Copy DOI

Abstract

Understanding the clonal architecture and evolutionary history of a tumour poses one of the key challenges to overcome treatment failure due to resistant cell populations. Previously, studies on subclonal tumour evolution have been primarily based on bulk sequencing and in some recent cases on single-cell sequencing data. Either data type alone has shortcomings with regard to this task, but methods integrating both data types have been lacking. Here, we present B-SCITE, the first computational approach that infers tumour phylogenies from combined single-cell and bulk sequencing data. Using a comprehensive set of simulated data, we show that B-SCITE systematically outperforms existing methods with respect to tree reconstruction accuracy and subclone identification. B-SCITE provides high-fidelity reconstructions even with a modest number of single cells and in cases where bulk allele frequencies are affected by copy number changes. On real tumour data, B-SCITE generated mutation histories show high concordance with expert generated trees.

Highlights

Understanding the clonal architecture and evolutionary history of a tumour poses one of the key challenges to overcome treatment failure due to resistant cell populations
We developed B-SCITE, a probabilistic approach for the inference of tumour mutation histories by the use of single-nucleotide variants (SNVs) data obtained from single-cell and bulk DNA sequencing (Fig. 2)
B-SCITE consists of a Markov chain Monte Carlo-based search scheme to traverse the space of possible tree topologies and a joint likelihood model for the evaluation of candidate trees based on their joint fit to the single-cell and bulk data

Summary

Introduction

Understanding the clonal architecture and evolutionary history of a tumour poses one of the key challenges to overcome treatment failure due to resistant cell populations. 1234567890():,; Cancer is a genetic disease that develops through a branched evolutionary process[1] It is characterised by the emergence of genetically distinct subclones through the random acquisition of mutations at the level of single cells and shifting prevalences at the subclone level through selective advantages purveyed by driver mutations. Most genetic analyses of tumours are currently based on nextgeneration sequencing data of bulk tumour samples Such data provide indirect measurements of the subclonal tumour composition in the form of aggregate total and variant read counts measured across hundreds of thousands or millions of cells. Sequencing multiple samples from the same tumour and increasing the coverage can to some extent mitigate these issues, but is not always practicable Another solution is the use of single-cell sequencing (SCS) data which provide mutation profiles of individual cells, such that the phylogeny can be directly inferred without any form of deconvolution. Classic approaches for phylogeny reconstruction are not suitable for dealing with these SCS-specific noise profiles, and a number of probabilistic approaches have been developed to account for the error types found in SCS data[28,29,30,31,32]

Objectives

Methods

Results

Conclusion