Parallel evolution is often assumed to result from repeated adaptation to novel, yet ecologically similar, environments. Here, we develop and analyse a mathematical model that predicts the probability of parallel genetic evolution from standing genetic variation as a function of the strength of phenotypic selection and constraints imposed by genetic architecture. Our results show that the probability of parallel genetic evolution increases with the strength of natural selection and effective population size and is particularly likely to occur for genes with large phenotypic effects. Building on these results, we develop a Bayesian framework for estimating the strength of parallel phenotypic selection from genetic data. Using extensive individual-based simulations, we show that our estimator is robust across a wide range of genetic and evolutionary scenarios and provides a useful tool for rigorously testing the hypothesis that parallel genetic evolution is the result of adaptive evolution. An important result that emerges from our analyses is that existing studies of parallel genetic evolution frequently rely on data that is insufficient for distinguishing between adaptive evolution and neutral evolution driven by random genetic drift. Overcoming this challenge will require sampling more populations and the inclusion of larger numbers of loci.

Viral phylogenies provide crucial information on the spread of infectious diseases, and many studies fit mathematical models to phylogenetic data to estimate epidemiological parameters such as the effective reproduction ratio (Re) over time. Such phylodynamic inferences often complement or even substitute for conventional surveillance data, particularly when sampling is poor or delayed. It remains generally unknown, however, how robust phylodynamic epidemiological inferences are, especially when there is uncertainty regarding pathogen prevalence and sampling intensity. Here we use recently developed mathematical techniques to fully characterize the information that can possibly be extracted from serially collected viral phylogenetic data, in the context of the commonly used birth-death-sampling model. We show that for any candidate epidemiological scenario, there exists a myriad of alternative, markedly different and yet plausible “congruent” scenarios that cannot be distinguished using phylogenetic data alone, no matter how large the dataset. In the absence of strong constraints or rate priors across the entire study period, neither maximum-likelihood fitting nor Bayesian inference can reliably reconstruct the true epidemiological dynamics from phylogenetic data alone; rather, estimators can only converge to the “congruence class” of the true dynamics. We propose concrete and feasible strategies for making more robust epidemiological inferences from viral phylogenetic data.

All species are locked in a continual struggle to adapt to local ecological conditions. In cases where species fail to locally adapt, they face reduced population growth rates, or even local extinction. Traditional explanations for limited local adaptation focus on maladaptive gene flow or homogeneous environmental conditions. These classical explanations have, however, failed to explain variation in the magnitude of local adaptation observed across taxa. Here we show that variable levels of local adaptation are better explained by trait dimensionality. First, we develop and analyse mathematical models that predict levels of local adaptation will increase with the number of traits experiencing spatially variable selection. Next, we test this prediction by estimating the relationship between dimensionality and local adaptation using data from 35 published reciprocal transplant studies. This analysis reveals a strong correlation between dimensionality and degree of local adaptation, and thus provides empirical support for the predictions of our model.

The results of genome-wide association studies are known to be affected by epistasis and gene-by-environment interactions. Using a statistical model....

Birth-death stochastic processes are the foundation of many phylogenetic models and are widely used to make inferences about epidemiological and macroevolutionary dynamics. There are a large number of birth-death model variants that have been developed; these impose different assumptions about the temporal dynamics of the parameters and about the sampling process. As each of these variants was individually derived, it has been difficult to understand the relationships between them as well as their precise biological and mathematical assumptions. Without a common mathematical foundation, deriving new models is non-trivial. Here we unify these models into a single framework, prove that many previously developed epidemiological and macroevolutionary models are all special cases of a more general model, and illustrate the connections between these variants. This unification includes both models where the process is the same for all lineages and those in which it varies across types. We also outline a straightforward procedure for deriving likelihood functions for arbitrarily complex birth-death(-sampling) models that will hopefully allow researchers to explore a wider array of scenarios than was previously possible. By re-deriving existing single-type birth-death sampling models we clarify and synthesize the range of explicit and implicit assumptions made by these models.

No abstract

Coevolutionary negative frequency dependent selection has been hypothesized to maintain genetic variation in host and parasites. Despite the extensive literature pertaining to host-parasite coevolution, the effect of matching-alleles (MAM) coevolution on the maintenance of genetic variation has not been explicitly modelled in a finite population. The dynamics of the MAM in an infinite population, in fact, suggests that genetic variation in these coevolving populations behaves neutrally. We find that while this is largely true in finite populations two additional phenomena arise. The first of these effects is that of coevolutionary natural selection on stochastic perturbations in host and pathogen allele frequencies. While this may increase or decrease genetic variation, depending on the parameter conditions, the net effect is small relative to that of the second phenomena. Following fixation in the pathogen, the MAM becomes one of directional selection, which in turn rapidly erodes genetic variation in the host. Hence, rather than maintain it, we find that, on average, matching-alleles coevolution depletes genetic variation.

Birth-death stochastic processes are the foundation of many phylogenetic models and are widely used to make inferences about epidemiological and macroevolutionary dynamics. There are a large number of birth-death model variants that have been developed; these impose different assumptions about the temporal dynamics of the parameters and about the sampling process. As each of these variants was individually derived, it has been difficult to understand the relationships between them as well as their precise biological and mathematical assumptions. And without a common mathematical foundation, deriving new models is non-trivial. Here we unify these models into a single framework, prove that many previously developed epidemiological and macroevolutionary models are all special cases of a more general model, and illustrate the connections between these variants. To do so, we develop a novel technique for deriving likelihood functions for arbitrarily complex birth-death(-sampling) models that will allow researchers to explore a wider array of scenarios than was previously possible. As an illustration of the utility of our mathematical approach, we use our approach to derive a yet unstudied variant of the birth-death process in which the key rates emerge deterministically from a classic susceptible infected recovered (SIR) epidemiological model.