Modelling the Population Genetics of Non-Equilibrium Molecular Evolution: Understanding The Forces that Shape The Genome

点击翻译按钮获取中文摘要

Thanks to transformational technological advances in the past decade, whole genome sequencing has become feasible in many species and populations (e.g., the Vertebrate Genomes Project is sequencing 70,000 vertebrate genomes). Such data resources present next-level opportunities to “read evolution's laboratory notebook”, and to make more powerful and richer inferences about patterns of genome sequence variations and their causes, about the natural history of selection, the sequence variations that work', and about the tree of life itself. However, substantial conceptual and computational challenges impede the full utilization of such data at present. Objectives The overarching aim of our research program is to understand, model, and infer the forces that shaped the evolution of vertebrate genomes. Significant recent effort has been expended in developing models of molecular evolution with an explicit population genetic basis, however, current models make a variety of simplifying assumptions inherited from theoretical population genetics that were popularized largely because they led to convenient mathematics in an era that predates modern computers (e.g., infinite sites and weak mutation). These assumptions may not always be justified in natural populations, and an important direction is thus to unshackle our models from such antiquated assumptions. Scientific Approach In a series of recent papers we demonstrated how general computational approaches for algebraically exact analysis of any Markov chain model in population genetics can be made feasible and highly efficient. These approaches allowed us to make a detailed theoretical analysis of the rate of substitution without classical assumptions, which showed that several predictions of the neutral theory of molecular evolution break down in populations with large but biologically relevant population mutation rates. In this research, we will determine if these predictions hold in natural populations. We will first translate our theoretical developments into models useful for analysing across-species sequence evolution, including time heterogeneity at both phylo- and population-genetic scales. We will use these and other approaches to test two main hypotheses: 1) The rate of substitution disobeys traditional weak-mutation theory when the population mutation rate parameter, theta, is as small as 0.05; and 2) Temporal variation in population size across a phylogeny can induce large-scale phylogenetic artifacts. Impact If validated, the impact of this work may be substantial for studies of molecular and genome evolution, phylogenetics, and in population and evolutionary genetics. In particular, nearly all methods for inferring the strength and direction of natural selection will need modification to apply to populations in vulnerable parameter ranges. We hope that the models developed will advance us along the path towards meaningfully more mechanistic models of molecular evolution.