Novel Multi-sample Scheme for Inferring Phylogenetic Markers from Whole Genome Tumor Profiles

Abstract

Computational cancer phylogenetics seeks to enumerate the temporal sequences of aberrations in tumor evolution, thereby delineating the evolution of possible tumor progression pathways, molecular subtypes and mechanisms of action. We previously developed a pipeline for constructing phylogenies describing evolution between major recurring cell types computationally inferred from whole-genome tumor profiles. The accuracy and detail of the phylogenies, however, depends on the identification of accurate, high-resolution molecular markers of progression, i.e., reproducible regions of aberration that robustly differentiate different subtypes and stages of progression. Here we present a novel hidden Markov model (HMM) scheme for the problem of inferring such phylogenetically significant markers through joint segmentation and calling of multi-sample tumor data. Our method classifies sets of genome-wide DNA copy number measurements into a partitioning of samples into normal (diploid) or amplified at each probe. It differs from other similar HMM methods in its design specifically for the needs of tumor phylogenetics, by seeking to identify robust markers of progression conserved across a set of copy number profiles. We show an analysis of our method in comparison to other methods on both synthetic and real tumor data, which confirms its effectiveness for tumor phylogeny inference and suggests avenues for future advances.

Reference-free inference of tumor phylogenies from single-cell sequencing data

“…The trees allow us to learn markers of progression driving key steps in tumor evolution, identify and classify tumor subtypes with possibly different underlying mechanisms of action, and enable predictive modeling of future stages of progression. Much prior work has shown the utility of tumor phylogenies using data from single-cell Fluorescent In-Situ Hybridization (FISH) [ 12 - 15 ] and microarray technologies [ 16 , 17 ]. Advances in deep sequencing technology, particularly single-cell sequencing [ 18 , 19 ], have extended this interest by promising a granular view of heterogeneity and progression within a single tumor.…”

A tutorial on the balanced minimum evolution problem

“…For example, phylogenies have been used to predict evolution of human influenza A [15,123]; to understand the relationships between the virulence and evolution of HIV [17,115,124,133]; to identify emerging viruses such as SARS [2,67,104]; to recreate and investigate ancestral proteins [33,160]; to design neuropeptides causing smooth muscle contraction [5]; to relate geographic patterns to ecological and macro-evolutionary processes [69,90,96,107]; or to uncover similarities in evolution of a number of human languages [12,79]. Phylogenies have also been used to study the evolutionary processes underlying the genetic factors involved in common human diseases [24,29,109,122,147,148] as well as those at the core of the progression of carcinomas over time [30,35,91,110,129,130,142,152]. In particular, in the cancer context, phylogenies allowed the remarkable classification of tumor cells of given pathologies in subfamilies characterized by specific evolutionary traits [142].…”

Reference-free inference of tumor phylogenies from single-cell sequencing data

“…The trees allow us to learn markers of progression driving key steps in tumor evolution, identify and classify tumor subtypes with possibly different underlying mechanisms of action, and enable predictive modeling of future stages of progression. Much prior work has shown the utility of tumor phylogenies using data from singlecell Fluorescent In-Situ Hybridization (FISH) [12][13][14][15] and microarray technologies [16,17]. Advances in deep sequencing technology, particularly single-cell sequencing [18,19], have extended this interest by promising a granular view of heterogeneity and progression within a single tumor.…”