Biology subjects

van Benthem, K. J.

Publications and source records attributed to van Benthem, K. J..

8 recordsLinked to original sources

bioRxiv2026

Identifying Evolutionary Relatedness Effects on Diversification from Phylogenies using Neural Networks

Reconstructing the forces that shaped macroevolutionary histories from extant phylogenies is fundamentally challenging: richly parameterized diversification models are often only weakly identifiable; different evolutionary mechanisms can yield nearly indistinguishable tree shapes. Here we use a model with evolutionary relatedness dependence to evaluate how much information about such forces can be recovered from simulated trees. We train graph neural networks and long short-term memory classifiers to distinguish three scenarios of feedback of diversity on diversification: effect of phylogenetic diversity (total branch length), evolutionary distinctiveness (average phylogenetic distance of a species to all other species in a clade), and nearest-neighbor distance (phylogenetic distance to the mostly closely related species). We also train a suite of regression networks to estimate the underlying diversification parameters. We then analyze classification performance, calibration of predicted class probabilities, regression errors, and their dependence on tree size and on the strength and sign of richness and relatedness effects. Across network architectures and complexity levels, scenario classification is only moderately accurate and strongly asymmetric as revealed by the confusion matrix. Trees generated under an effect of nearest-neighbor distance on diversification tend to be correctly classified, whereas those with an effect of evolutionary distinctiveness are frequently misclassified. Regression networks systematically shrink predictions toward the empirical mean, especially for complex models, suggesting broad regions of parameter space with low identifiability. Strong global dependence of diversification rates on diversity further erodes recoverability by driving large variations in tree size that mask the subtler signatures of related-ness effects. In contrast, sufficiently strong speciation-relatedness effects can carve out narrow regions of parameter space in which scenarios and parameters become practically recoverable. Together, our results provide a map of when neural networks can and cannot infer diversification mechanisms from extant trees under our evolutionary relatedness dependence model, and they underscore the need for additional data or constraints when using flexible diversification models for macroevolutionary inference.

evolutionary biology↗

bioRxiv2025

Detecting Nuclear Pore Complex assembly in living cells

The formation of nuclear pore complexes (NPCs) -- vital gateways regulating nuclear-cytoplasmic transport -- is a highly orchestrated process requiring the integration of hundreds of nucleoporins into the nuclear envelope. A major challenge in studying this assembly process in living cells has been the difficulty to distinguish newly forming NPCs from their mature counterparts. Here, we present a powerful nanobody-based approach that overcomes this limitation. We demonstrate that a nanobody targeting the nucleoporin Nic96 from Saccharomyces cerevisiae selectively binds newly synthesized Nic96 subcomplexes prior to its incorporation into NPCs in vivo. Importantly, nanobody-bound Nic96 is incorporated in NPCs, and expression of the nanobody does not disrupt nuclear transport, cell growth, or lifespan, nor does it show genetic interactions with known NPC assembly surveillance pathways -- making it an ideal, non-perturbing tool to study NPC biogenesis in yeast. Illustrating the use of the Nic96 nanobody we report novel aspects of the early stages of assembly, including co-recruitment of Kap121 and VHH[Nic96] to putative assembly sites. In addition, we show local enrichment of newly synthesized nucleoporins on nuclear envelope proximal lipid droplets, including a specific subset of Pdr16- and Ldo16-positive lipid droplets near the nucleus-vacuole junction. When NPC assembly is delayed, association of Nic96 subcomplexes with lipid droplets increases. These findings illustrate how the strategy to pulse-label newly forming NPCs opens new avenues for dissecting the spatiotemporal regulation of NPC assembly and misassembly.

cell biology↗

bioRxiv2025

Can the marginal male hypothesis explain spatial density variation in pinnipeds?

The marginal male hypothesis is a mechanism proposed to explain gregariousness, for example in pinnipeds. Here, we explore whether this mechanism, combined with density-dependent pup survival, can also account for heterogeneity in density across colonies, as observed for example in Antarctic fur seals (Arctocephalus gazella). We build a discrete-time matrix model inspired by the Antarctic fur seal to investigate how differences in density between two colonies can emerge through mate choice and density-dependent pup mortality. Our model assumes a heritable male colony preference that is coupled to competitive ability, i.e. a local siring advantage. Female colony preference is modelled as an independent trait that is allowed to evolve separately. Pup survival decreases with local density. Based on numerical analysis of the model, we find slight differences in density between the colonies at equilibrium. When siring advantage is high, and the survival penalty for males with siring advantage is low, the colony that is preferred by the high-quality males actually becomes the less dense colony. Our model serves as a proof of concept that such density differences can emerge. We expect that by including additional environmental factors, future models will explain a larger part of the observed variation in density.

ecology↗

bioRxiv2023

Multi-scale Effects of Habitat Loss and the Role of Trait Variation

Habitat loss (HL) is a major cause of species extinctions. Although effects of HL beyond the directly impacted area have been previously observed, they are not very well understood, especially in an eco-evolutionary context. To start filling this gap, we study a two-patch deterministic consumer-resource model, with one of the patches experiencing loss of resources. Our model allows foraging and mating within a patch as well as between patches. We then introduce heritable variation in consumer traits to investigate eco-evolutionary dynamics and compare results with constant or no trait variation scenarios. Our results show that HL indeed reduces consumer densities in the neighboring patch, but when the resources are overexploited, HL in one patch can increase the consumer densities in the neighbouring patch. Yet at the landscape scale, the effect of HL on consumer densities is consistently negative. In presence of HL, patch isolation has positive effects on consumer density in the patch experiencing HL and mostly negative effects on the neighbouring patch. The landscape level pattern depends on which of these effects are dominant at the local scale. Evolution always increased resistance of consumers in the affected patch to HL, with varied effects at the landscape level. Finally, we also show a possibility of landscape level consumer extinction due to HL in a local patch when the cross-patch dependence is high, and foraging and mating preferences are coupled. Eco-evolutionary dynamics can rescue consumers from such extinction in some cases if their death rates are sufficiently small. Our findings show that HL at a local scale can affect the neighbouring patch and the landscape as a whole, and that heritable trait variation can provide some resistance against HL. We thus suggest joint consideration of multiple spatial scales and trait variation when assessing and predicting the impacts of HL.

ecology↗

bioRxiv2021

Nonlinear averaging in 2D: intraspecific variation in two interacting species and consequences for population dynamics

Interactions between two species, e.g. between a predator species and a prey species, can often be described as the sum of many individual-by-individual interactions whose outcomes depend on the traits of the interacting individuals. These traits often vary substantially among individuals in each species, and individuals do not always interact randomly, e.g. due to plastic responses to a shared environmental factor in a heterogeneous landscape. Here we investigate the impact of intraspecific trait variation (ITV) and such interspecific trait correlations on species interactions via nonlinear averaging. Building on past models that integrate over an interaction kernel to obtain the impacts of ITV, we develop a modeling framework that allows to model arbitrary species interactions, with interspecific trait correlations as novel feature. Based on two key ingredients, a joint trait distribution and a two-dimensional interaction function, the average interaction parameters (e.g. average predation rate) can be quantified numerically, approximated using an insightful Taylor approximation, and compared to cases without ITV. We highlight two applications of our framework. First, we study the quantitative and qualitative effects of ITV and trait correlations in a simple predator-prey model and show that even in the absence of evolution, variation and trait correlations among interacting individuals can make or break the coexistence between species. Second, we use simulated field data for a predator-prey system to show how the impact of ITV on an ecological interaction can be estimated from empirical data. HighlightsO_LIWe model how intraspecific trait variation in two species affects their interaction. C_LIO_LIOur framework allows correlations between the traits of interacting individuals. C_LIO_LITrait variation and correlations can strongly affect ecological interactions. C_LIO_LITrait variation and correlations can make or break coexistence. C_LIO_LIThe effect of intraspecific trait variation can be estimated from data. C_LI

ecology↗

bioRxiv2020

Waiting for love but not forever: modelling the evolution of waiting time to selfing in hermaphrodites

Although mixed mating systems involving both selfing and outcrossing are fairly common in hermaphrodites, it is unclear how they are maintained. In some species, individuals delay self-fertilization while they have not found a mating partner. The waiting time is subject to two opposing selection pressures: waiting helps to avoid inbreeding depression in offspring by increasing the density-dependent probability to encounter a mate, but also increases the risk of dying before reproduction. In some species waiting time can vary between individuals and be heritable. We therefore used an individual-based model to explore how delayed selfing evolves in response to density and density fluctuations. We find that at high density, when individuals meet often, drift drives waiting time; at intermediate densities, strong inbreeding depression causes waiting time to increase; and at low densities, inbreeding depression is purged, and waiting time approaches zero. Positive feedback loops drive the system to either complete selfing or complete outcrossing. Fluctuating density can slow down convergence to these alternative stable states. However, mixed mating, in the sense of either a stable polymorphism in waiting times, or stable intermediate times, was never observed. Thus, additional factors need to be explored to explain the persistence of delayed selfing.

evolutionary biology↗

bioRxiv2019

Density dependence on multiple spatial scales maintains spatial variation in both abundance and traits

Population density affects fitness through various processes, such as mate finding and competition. The fitness of individuals in a population can in turn affect its density, making population density a key quantity linking ecological and evolutionary processes. Density effects are, however, rarely homogeneous. Different life-history processes can be affected by density over different spatial scales. In birds, for example, competition for food may depend on the number of birds nesting in the direct vicinity, while competition for nesting sites may occur over larger areas. Here we investigate how the effects of local density and of density in nearby patches can jointly affect the emergence of spatial variation in abundance as well as phenotypic diversification. We study a two-patch model that is described by coupled ordinary differential equations. The patches have no intrinsic differences: they both have the same fitness function that describes how an individuals fitness depends on density in its own patch as well as the density in the other patch. We use a phase-space analysis, combined with a mathematical stability analysis to study the long-term behaviour of the system. Our results reveal that the mutual effect that the patches have on each other can lead to the emergence and long-term maintenance of a low and a high density patch. We then add traits and mutations to the model and show that different selection pressures in the high and low density patch can lead to diversification between these patches. Via eco-evolutionary feedbacks, this diversification can in turn lead to changes in the long-term population densities: under some parameter settings, both patches reach the same equilibrium density when mutations are absent, but different equilibrium densities when mutations are allowed. We thus show how, even in the absence of differences between patches, interactions between them can lead to differences in long-term population density, and potentially to trait diversification.

evolutionary biology↗

bioRxiv2017

Modeling Zero-Inflated Count Data With glmmTMB

Ecological phenomena are often measured in the form of count data. These data can be analyzed using generalized linear mixed models (GLMMs) when observations are correlated in ways that require random effects. However, count data are often zero-inflated, containing more zeros than would be expected from the standard error distributions used in GLMMs, e.g., parasite counts may be exactly zero for hosts with effective immune defenses but vary according to a negative binomial distribution for non-resistant hosts.\n\nWe present a new R package, glmmTMB, that increases the range of models that can easily be fitted to count data using maximum likelihood estimation. The interface was developed to be familiar to users of the lme4 R package, a common tool for fitting GLMMs. To maximize speed and flexibility, estimation is done using Template Model Builder (TMB), utilizing automatic differentiation to estimate model gradients and the Laplace approximation for handling random effects. We demonstrate glmmTMB and compare it to other available methods using two ecological case studies.\n\nIn general, glmmTMB is more flexible than other packages available for estimating zero-inflated models via maximum likelihood estimation and is faster than packages that use Markov chain Monte Carlo sampling for estimation; it is also more flexible for zero-inflated modelling than INLA, but speed comparisons vary with model and data structure. Our package can be used to fit GLMs and GLMMs with or without zero-inflation as well as hurdle models. By allowing ecologists to quickly estimate a wide variety of models using a single package, glmmTMB makes it easier to find appropriate models and test hypotheses to describe ecological processes.

ecology↗