Utilizing developmental dynamics for evolutionary prediction and control.

Understanding, predicting, and controlling the phenotypic consequences of genetic and environmental change is essential to many areas of fundamental and applied biology. In evolutionary biology, the generative process of development is a major source of organismal evolvability that constrains or facilitates adaptive change by shaping the distribution of phenotypic variation that selection can act upon. While the complex interactions between genetic and environmental factors during development may appear to make it impossible to infer the consequences of perturbations, the persistent observation that many perturbations result in similar phenotypes indicates that there is a logic to what variation is generated. Here, we show that a general representation of development as a dynamical system can reveal this logic. We build a framework that allows predicting the phenotypic effects of perturbations, and conditions for when the effects of perturbations of different origins are concordant. We find that this concordance is explained by two generic features of development, namely the dynamical dependence of the phenotype on itself and the fact that all perturbations must affect the developmental process to have an effect on the phenotype. We apply our theoretical framework to classical models of development and show that it can be used to predict the evolutionary response to selection using information of plasticity and to accelerate evolution in a desired direction. The framework we introduce provides a way to quantitatively interchange perturbations, opening an avenue of perturbation design to control the generation of variation.

Twenty years on from Developmental Plasticity and Evolution: middle-range theories and how to test them.

In Developmental Plasticity and Evolution, Mary-Jane West-Eberhard argued that the developmental mechanisms that enable organisms to respond to their environment are fundamental causes of adaptation and diversification. Twenty years after publication of this book, this once so highly controversial claim appears to have been assimilated by a wealth of studies on 'plasticity-led' evolution. However, we suggest that the role of development in explanations for adaptive evolution remains underappreciated in this body of work. By combining concepts of evolvability from evolutionary developmental biology and quantitative genetics, we outline a framework that is more appropriate to identify developmental causes of adaptive evolution. This framework demonstrates how experimental and comparative developmental biology and physiology can be leveraged to put the role of plasticity in evolution to the test.

A data-driven framework to model the organism-environment system.

Organisms modify their development and function in response to the environment. At the same time, the environment is modified by the activities of the organism. Despite the ubiquity of such dynamical interactions in nature, it remains challenging to develop models that accurately represent them, and that can be fitted using data. These features are desirable when modeling phenomena such as phenotypic plasticity, to generate quantitative predictions of how the system will respond to environmental signals of different magnitude or at different times, for example, during ontogeny. Here, we explain a modeling framework that represents the organism and environment as a single coupled dynamical system in terms of inputs and outputs. Inputs are external signals, and outputs are measurements of the system in time. The framework uses time-series data of inputs and outputs to fit a nonlinear black-box model that allows to predict how the system will respond to novel input signals. The framework has three key properties: it captures the dynamical nature of the organism-environment system, it can be fitted with data, and it can be applied without detailed knowledge of the system. We study phenotypic plasticity using in silico experiments and demonstrate that the framework predicts the response to novel environmental signals. The framework allows us to model plasticity as a dynamical property that changes in time during ontogeny, reflecting the well-known fact that organisms are more or less plastic at different developmental stages.

A method to predict the response to directional selection using a Kalman filter.

Predicting evolution remains challenging. The field of quantitative genetics provides predictions for the response to directional selection through the breeder's equation, but these predictions can have errors. The sources of these errors include omission of traits under selection, inaccurate estimates of genetic variance, and nonlinearities in the relationship between genetic and phenotypic variation. Previous research showed that the expected value of these prediction errors is often not zero, so predictions are systematically biased. Here, we propose that this bias, rather than being a nuisance, can be used to improve the predictions. We use this to develop a method to predict evolution, which is built on three key innovations. First, the method predicts change as the breeder's equation plus a bias term. Second, the method combines information from the breeder's equation and from the record of past changes in the mean to predict change using a Kalman filter. Third, the parameters of the filter are fitted in each generation using a learning algorithm on the record of past changes. We compare the method to the breeder's equation in two artificial selection experiments, one using the wing of the fruit fly and another using simulations that include a complex mapping of genotypes to phenotypes. The proposed method outperforms the breeder's equation, particularly when traits under selection are omitted from the analysis, when data are noisy, and when additive genetic variance is estimated inaccurately or not estimated at all. The proposed method is easy to apply, requiring only the trait means over past generations.

Evolution of the G Matrix under Nonlinear Genotype-Phenotype Maps.

AbstractThe G matrix is a statistical summary of the genetic basis of a set of traits and a central pillar of quantitative genetics. A persistent controversy is whether G changes slowly or quickly over time. The evolution of G is important because it affects the ability to predict, or reconstruct, evolution by selection. Empirical studies have found mixed results on how fast G evolves. Theoretical work has largely been developed under the assumption that the relationship between genetic variation and phenotypic variation-the genotype-phenotype map (GPM)-is linear. Under this assumption, G is expected to remain constant over long periods of time. However, according to developmental biology, the GPM is typically complex and nonlinear. Here, we use a GPM model based on the development of a multicellular organ to study how G evolves. We find that G can change relatively fast and in qualitative different ways, which we describe in detail. Changes can be particularly large when the population crosses between regions of the GPM that have different properties. This can result in the additive genetic variance in the direction of selection fluctuating over time and even increasing despite the eroding effect of selection.

Is evolution predictable? Quantitative genetics under complex genotype-phenotype maps.

A fundamental aim of post-genomic 21st century biology is to understand the genotype-phenotype map (GPM) or how specific genetic variation relates to specific phenotypic variation. Quantitative genetics approximates such maps using linear models, and has developed methods to predict the response to selection in a population. The other major field of research concerned with the GPM, developmental evolutionary biology, or evo-devo, has found the GPM to be highly nonlinear and complex. Here, we quantify how the predictions of quantitative genetics are affected by a complex, nonlinear map based on the development of a multicellular organ. We compared the predicted change in mean phenotype for a single generation using the multivariate breeder's equation, with the change observed from the model of development. We found that there are frequent disagreements between predicted and observed responses to selection due to the nonlinear nature of the genotype-phenotype map. Our results are a step toward integrating the fields studying the GPM.