A mechanistically interpretable model of the retinal neural code for natural scenes with multiscale adaptive dynamics

The visual system processes stimuli over a wide range of spatiotemporal scales, with individual neurons receiving input from tens of thousands of neurons whose dynamics range from milliseconds to tens of seconds. This poses a challenge to create models that both accurately capture visual computations and are mechanistically interpretable. Here we present a model of salamander retinal ganglion cell spiking responses recorded with a multielectrode array that captures natural scene responses and slow adaptive dynamics. The model consists of a three-layer convolutional neural network (CNN) modified to include local recurrent synaptic dynamics taken from a linear-nonlinear-kinetic (LNK) model \cite{ozuysal2012linking}. We presented alternating natural scenes and uniform field white noise stimuli designed to engage slow contrast adaptation. To overcome difficulties fitting slow and fast dynamics together, we first optimized all fast spatiotemporal parameters, then separately optimized recurrent slow synaptic parameters. The resulting full model reproduces a wide range of retinal computations and is mechanistically interpretable, having internal units that correspond to retinal interneurons with biophysically modeled synapses. This model allows us to study the contribution of model units to any retinal computation, and examine how long-term adaptation changes the retinal neural code for natural scenes through selective adaptation of retinal pathways.

Attracting pollinators vs escaping herbivores: eco-evolutionary dynamics of plants confronted with an ecological trade-off

A large number of plant traits are subject to an ecological trade-off between attracting pollinators and escaping herbivores. The interplay of both plant-animal interactions determines their evolution. Within a plant-pollinator-herbivore community in which interaction strengths depend on trait-matching, eco-evolutionary dynamics are studied using the framework of adaptive dynamics. We characterize the type of selection acting on the plant phenotype and the consequences for multispecies coexistence. We find that pollination favors stabilizing selection and coexistence. In contrast, herbivory fosters runaway selection, which threatens plant-animal coexistence. These contrasting dynamics highlight the key role of ecological trade-offs in structuring ecological communities. In particular, we show that disruptive selection is possible when such trade-offs are strong. While the interplay of pollination and herbivory is known to maintain plant polymorphism in several cases, our work suggests that it might also have fueled the diversification process itself.

Quantifying the effect of genetic, environmental and individual demographic stochastic variability for population dynamics in Plantago lanceolata

Simple demographic events, the survival and reproduction of individuals, drive population dynamics. These demographic events are influenced by genetic and environmental parameters, and are the focus of many evolutionary and ecological investigations that aim to predict and understand population change. However, such a focus often neglects the stochastic events that individuals experience throughout their lives. These stochastic events also influence survival and reproduction and thereby evolutionary and ecological dynamics. Here, we illustrate the influence of such non-selective demographic variability on population dynamics using population projection models of an experimental population of Plantago lanceolata. Our analysis shows that the variability in survival and reproduction among individuals is largely due to demographic stochastic variation with only modest effects of differences in environment, genes, and their interaction. Common expectations of population growth, based on expected lifetime reproduction and generation time, can be misleading when demographic stochastic variation is large. Large demographic stochastic variation exhibited within genotypes can lower population growth and slow evolutionary adaptive dynamics. Our results accompany recent investigations that call for more focus on stochastic variation in fitness components, such as survival, reproduction, and functional traits, rather than dismissal of this variation as uninformative noise.

Spatial social dilemmas promote diversity

Cooperative investments in social dilemmas can spontaneously diversify into stably coexisting high and low contributors in well-mixed populations. Here we extend the analysis to emerging diversity in (spatially) structured populations. Using pair approximation, we derive analytical expressions for the invasion fitness of rare mutants in structured populations, which then yields a spatial adaptive dynamics framework. This allows us to predict changes arising from population structures in terms of existence and location of singular strategies, as well as their convergence and evolutionary stability as compared to well-mixed populations. Based on spatial adaptive dynamics and extensive individual-based simulations, we find that spatial structure has significant and varied impacts on evolutionary diversification in continuous social dilemmas. More specifically, spatial adaptive dynamics suggests that spontaneous diversification through evolutionary branching is suppressed, but simulations show that spatial dimensions offer new modes of diversification that are driven by an interplay of finite-size mutations and population structures. Even though spatial adaptive dynamics is unable to capture these new modes, they can still be understood based on an invasion analysis. In particular, population structures alter invasion fitness and can open up new regions in trait space where mutants can invade, but that may not be accessible to small mutational steps. Instead, stochastically appearing larger mutations or sequences of smaller mutations in a particular direction are required to bridge regions of unfavorable traits. The net effect is that spatial structure tends to promote diversification, especially when selection is strong.

A general theoretical framework for trait-based eco-evolutionary dynamics: population structure, intraspecific variation, and community assembly

To understand how functional traits shape ecological communities it is necessary to understand both how traits across the community affect its functioning and how eco-evolutionary dynamics within the community change the traits over time. Of particular interest are so-called evolutionarily stable communities (ESCs), since these are the end points of eco-evolutionary dynamics and can persist over long time scales. One theoretical framework that has successfully been used for assembling ESCs is adaptive dynamics. However, this framework cannot account for intraspecific variation---neither locally nor across structured populations. On the other hand, in moment-based approaches, intraspecific variation is accommodated, but community assembly has been neglected. This is unfortunate as some questions regarding for example local adaptation vis-a-vis diversification into multiple species requires both facets. In this paper we develop a general theoretical framework that bridges the gap between these two approaches. We showcase how ESCs can be assembled using the framework, and illustrate various aspects of the framework using two simple models of resource competition. We believe this unifying framework could be of great use to address questions regarding the role of functional traits in communities where population structure, intraspecific variation, and eco-evolutionary dynamics are all important.

Inevitability of Red Queen evolution driven by organismic complexity and simple feedback via environmental modification

Evolution in complex high-dimensional phenotype spaces can be very different than the caricature of uphill evolutionary trajectories in a low-dimensional fitness landscape. And slight modifications of the environment can have large consequences for the future evolution. Here, the simplest approximation of evolution, an almost-always clonal population evolving by small effect mutations under deterministic "adaptive" dynamics, is studied. The complexities of organisms and their interactions with their environments are caricatured by population growth rates being smoothly varying random functions in very high dimensional phenotype spaces. In a fixed environment, there are huge numbers of fitness maxima, yet evolutionary trajectories wander around amongst saddles, gradually slowing down but still wandering widely and without committing to any maximum. But with even very small changes of the environment caused by the phenotypic changes, after an initial transient the evolution continues forever without further slowing down. In this Red Queen "phase" the apparent rate of increase of the fitness saturates (at a feedback strength-dependent rate) and the trajectories perpetually wander over large phenotypic distances. Organismic complexities, caricatured by a large number of constraints on the molecular-level phenotype, together with the simplest possible interactions of the organisms with their environment, are shown to be sufficient to cause such Red Queen dynamics. Arguments are made for the ubiquity of such behavior.

Prevalence of mutualism in a simple model of microbial co-evolution

Evolutionary transitions among ecological interactions are widely known, although their detailed dynamics remain absent for most population models. Adaptive dynamics has been used to illustrate how the parameters of population models might shift through evolution but within an ecological regime. Here we use adaptive dynamics combined with a generalised logistic model of population dynamics to show that transitions of ecological interactions might appear as a consequence of evolution. To this purpose, we introduce a two-microbial toy model in which population parameters are determined by a bookkeeping of resources taken from (and excreted to) the environment, as well as from the byproducts of the other species. Despite its simplicity, this model exhibits all sorts of ecological transitions, some of which resemble those found in nature. Overall, the model shows a clear trend toward the emergence of mutualism.

Interplay between extreme drift and selection intensities favors the fixation of beneficial mutations in selfing maize populations

Population and quantitative genetic models provide useful approximations to predict long-term selection responses sustaining phenotypic shifts, and underlying multilocus adaptive dynamics. Valid across a broad range of parameters, their use for understanding the adaptive dynamics of small selfing populations undergoing strong selection intensity (thereafter High Drift-High selection regime, HDHS) remains to be explored. Saclay Divergent Selection Experiments (DSEs) on maize flowering time provide an interesting example of populations evolving under HDHS, with significant selection responses over 20 generations in two directions. We combined experimental data from Saclay DSEs, forward individual-based simulations, and theoretical predictions to dissect the evolutionary mechanisms at play in the observed selection responses. We asked two main questions: How do mutations arise, spread, and reach fixation in populations evolving under HDHS? How does the interplay between drift and selection influence observed phenotypic shifts? We showed that the long-lasting response to selection in small populations is due to the rapid fixation of mutations occurring during the generations of selection. Among fixed mutations, we also found a clear signal of enrichment for beneficial mutations revealing a limited cost of selection. Both environmental stochasticity and variation in selection coefficients likely contributed to exacerbate mutational effects, thereby facilitating selection grasp and fixation of small-effect mutations. Together our results highlight that despite a small number of polymorphic loci expected under HDHS, adaptive variation is continuously fueled by a vast mutational target. We discuss our results in the context of breeding and long-term survival of small selfing populations.