Choose timezone
Your profile timezone:
Powered by Indico
v3.3.5
IMPORTANT DATES
Workshop: 30 June – 3 July, 2025
The workshop starts on June 30 at 5pm and ends on July 3 in the evening.
Pre-school: June 29 (afternoon) & June 30, 2025
Registration deadline: 15 March, 2025 (extended until 31 March for late registrations)
Notification of acceptance: 22 March, 2025
OVERVIEW
Evolutionary rescue is a topic of great interest, from medicine to agriculture to conservation, and from natural observations to experiments to theory. In 2023, we organised a workshop on ‘Mathematical models of evolutionary rescue’. In this second edition, we now aim to bring modelers and experimentalists together to discuss (1) which theoretical insights should be tested experimentally and how this could be done, (2) which experimental observations could be clarified with the help of theoretical models, and (3) which open questions could be addressed by the co-development of models and experiments. In addition to invited and contributed talks and posters there will be discussion sessions, designed to increase cross-talk between theory and experiments.
INVITED SPEAKERS
Helen Alexander (University of Edinburgh)
Lutz Becks (University of Konstanz)
Robert D. Holt (University of Florida)
Jitka Polechová (University of Vienna)
Laure Olazcuaga (CNRS - SETE)
PRE-SCHOOL
In addition to the workshop, we offer a ‘pre-school’ prior to the workshop, in which we provide an introduction to models of evolutionary rescue (afternoon of June 29 – June 30). This school will be open for workshop participants (e.g. students or experimental researchers who would like to gain better insights into modeling) and members of the institute. The pre-school will take place if there is sufficient interest.
APPLICATION & REGISTRATION
We hope that all participants present a talk or poster but it is not mandatory. However, if you do not submit an abstract please briefly state your motivation for participating (max. 250 words) in the appropriate field. Applicants will be notified about acceptance by March 22.
Registration is free. However, you need to pay for your own travel and accommodation. If you do not have sufficient funds to cover these expenses, please provide some information (including estimated gaps in funding) in the respective field upon registration. We may be able to subsidize travel in
exceptional cases.
CONTACT
For all administrative queries (including invitation letters for visa applications), please contact Maren Lehmann (lehmann@evolbio.mpg.de).
If you have any other questions, please contact the organizers, Hildegard Uecker (uecker@evolbio.mpg.de) and Matthew Osmond (mm.osmond@utoronto.ca).
Dinner at Indian Restaurant Taj Mahal directly opposite the Pink Residence Building on Rautenbergstrasse
The Luria Delbrück experiment has 80 years of history and is arguably the most widely used method of analyzing evolutionary rescue (ER) in microbes. It consists in growing replicate, initially clonal, populations in permissive conditions then abruptly switching each replicate onto a petri dish containing selective lethal conditions (eg antibiotics or lytic phages). Finally, the number of growing colonies on each dish (evolutionary rescues) is counted and analyzed using the so-called Luria Delbrück distribution. The model behind this analysis is de facto assuming that only mutations preexiting the switch to stress are causal of visible colonies, so it is a purely standing variance ER model. It also typically assumes a single mutant type, while it is well known and observed that ER can mobilize a diversity of genetic bases, especially at low lethal doses. Here I will present preliminary work on a possible modification of the protocol to estimate both de novo and standing contributions, and to allow for a distribution of type demographic traits. I will do so in the context of a Feller diffusion approximation, which, to my knowledge, has not been used in the context of the Luria Delbrück theory or data analysis. This approximation provides relatively handy simplifications for a diversity of bacterial demographies. Any preliminary result on the statistical properties of estimators will be dealt with only if time allows.
Stochastic establishment of adaptive genotypes (survival and outgrowth of initially rare lineages) is a key step in evolutionary rescue. Despite featuring in theoretical models, this process has received little attention in experimental biology. Here I will describe how we can quantify the probability of establishment with low-tech, high-throughput experimental assays combined with statistical inference. I will present an application to establishment of resistant bacteria (Pseudomonas aeruginosa) under antibiotic treatment, an example of evolutionary rescue in medical contexts. Our experiments support the theoretical expectation that, despite a positive average population growth rate, establishment from single individuals frequently fails. Moreover, we find that the surrounding, initially large, antibiotic-sensitive population can either inhibit or (surprisingly) facilitate establishment of resistance. More broadly, bacteria could provide a relevant experimental system for studying the role of biotic interactions in evolutionary rescue.
Time-kill experiments are commonly used to characterise the pharmacodynamic properties of antibiotics. During these experiments, a bacterial population is exposed to a specific concentration of an antibiotic over time, typically lasting 24 hours. Given high enough concentrations, a decrease of the bacterial population can be observed, which is used to calculate the kill rate constant for the specific concentration. Unfortunately, evolution cannot simply be “turned off” during these experiments and sometimes regrowth of the population occurs after an initial phase of rapid killing, resulting in the well-known U-shaped curve of evolutionary rescue. During our experiments to characterise the pharmacodynamic properties of fosfomycin, amikacin and their combination for different E. coli strains, regrowth was observed for every concentration (even for the combination). In order to properly assess the pharmacodynamics of these drugs, knowledge on the underlying cause of this short-term rescue is needed. With a mathematical model, we aim to investigate different hypothesis on the underlying reason of the regrowth behaviour. Will this be an example for evolutionary rescue due to de novo evolution or might it result from the initial heterogeneity of the population with respect to antibiotic susceptibility?
The evolution of resistance is a major challenge in medicine and agriculture. Organisms can evolve resistance through a variety of mechanisms, including target modification, metabolic changes, and efflux systems. In diploid or polyploid organsims, the dominance of resistance mutations has a strong influence on the dynamics of resistance evolution. The dominance of an allele can differ across environmental conditions and may depend on the pesticide or the drug dose. A particularly intriguing phenomenon is dominance reversal, where resistance alleles are recessive or dominant depending on the environment. This poster provides an overview of common resistance mechanisms, explains the concept of dominance reversal, and discusses its potential implications for the evolution and management of resistance. By highlighting both classical and emerging perspectives, this work aims to stimulate discussion on how integrating context-dependent dominance can improve our understanding of resistance evolution in diploid (and polyploid) organisms and inform more effective intervention strategies.
While different stages of mutualism can be observed in natural communities, the dynamics and mechanisms underlying the gradual erosion of independence of the initially autonomous organisms are not yet fully understood. In this study, by conducting the laboratory evolution on an engineered microbial community, we reproduce and molecularly track the stepwise progression towards enhanced partner entanglement. We observe that the evolution of the community both strengthens the existing metabolic interactions and leads to the emergence of de novo interdependence between partners for nitrogen metabolism, which is a common feature of natural symbiotic interactions. Selection for enhanced metabolic entanglement during the community evolution repeatedly occurred indirectly, via pleiotropies and trade-offs within cellular regulatory networks, and with no evidence of group selection. The indirect positive selection of metabolic dependencies between microbial community members, which results from the direct selection of other coupled traits in the same regulatory network, may therefore be a common but underappreciated driving force guiding the evolution of natural mutualistic communities.
It became clear shortly after the discovery of the first antibiotics that bacteria are able to survive and evade antibiotic treatment. Ever since, understanding the emergence of antimicrobial resistance is of great interest and a large number of studies have addressed the mechanisms of bacterial evolution in the presence of antibiotics. However, such studies often focus exclusively on antibiotic resistance, while other effects such as tolerance and persistence evolution remain poorly understood. Here, we present a theoretical framework to describe the decline and recovery of a bacterial population under periodic antibiotic stress. Based on a well established model of Dose Response Curves, we define an effective fitness measure that quantifies the net growth rate per cycle. We show that tolerant mutants, characterized by a lower death rate, can significantly increase the bacterial fitness and potentially threaten the treatment efficacy, even if no change in resistance is involved. Based on the effective fitness model, we generalize the concept of trade-off induced fitness landscapes to the trade-off between growth and death rates, and study the evolution of tolerance as a random adaptive walk in this two-dimensional phenotype space.
The recurrent exposure to herbicides in agricultural landscapes forces weeds to adapt in a race against extinction. What role newly arising mutations and pre-existing variation play in this evolution of herbicide resistance is critical for developing management strategies. In this talk, I will present a multitype Galton-Watson process model of rapid adaptation in response to strong selection, capturing complex life cycles of sexual and asexual reproduction and dormancy. Applying our approach to herbicide resistance evolution in a perennial weed, I will derive the probability of resistance causing treatment failure and the waiting time distribution until resistant plants establish. Finally, I will illustrate the effect of self-pollination in influencing the probability and timing of resistance adaptation and the primary source of adaptive variation.
Theoretical models suggest that when the environment fluctuates slowly over hundreds of generations, populations primarily adapt to the environmental optimum through genetic changes rather than relying on phenotypic plasticity. Experimental evolution shows that gradual environmental change promotes the accumulation of smaller-effect mutations, leading to greater fitness than adapting to a single drastic environmental change. Long-term studies confirm continuous mutation accumulation over time. Here we conduct a long-term evolutionary experiment with slowly fluctuating environments using experimental populations of fission yeast founded from a single clone. We gradually introduce and intensify the stress in the environment, then return equally slowly to the initial conditions without the stressor. By sequencing the populations, we will determine whether adaptation occurs primarily through genetic changes or phenotypic plasticity. We predict that when the environment starts to change back, instead of reverting their changes, compensatory mutations that eliminate any possible trade-offs with the original environment will be fixed. When compared to populations evolved under constant conditions, we can determine whether environmental fluctuations promote genetic divergence. To test whether accumulated mutations actively contribute to adaptation, we will compare the growth rates of evolved and ancestral populations across different environments. Our findings will reveal whether environmental fluctuations promote genetic divergence and compensatory adaptation, shedding light on how populations genetically respond to slow and reversible environmental change.
Host-associated microbiomes are increasingly recognized as key players in adaptive evolution, shaping host responses to environmental stressors and exhibiting varying degrees of heritability across generations. Given their potential to influence host fitness, microbiomes may play a crucial role in facilitating evolutionary rescue. However, the conditions under which microbiomes promote evolutionary rescue remain poorly understood. To address this, we develop a mathematical model that integrates microbiome-mediated evolution with established evolutionary rescue theory, explicitly incorporating different modes of microbial transmission. Our results reveal that microbiome heritability indices alone provide limited predictive power for microbiome-mediated evolutionary rescue. Instead, we identify specific transmission mechanisms that govern the likelihood of host population persistence. This finding underscores microbiomes as key but often overlooked determinants of adaptive potential in changing environments. By linking microbial ecology with evolutionary theory, our framework provides new insights into how microbiomes buffer host populations against environmental stressors. We discuss implications for conservation biology, agriculture, and human health, highlighting how targeted microbial interventions could enhance population resilience and adaptive capacity.
Treatment of urinary tract infections (UTIs) and the prevention of their recurrence is a pressing global health problem. In a UTI, pathogenic bacteria not only reside in the bladder lumen but also attach to and invade the bladder tissue. Planktonic, attached, and intracellular bacteria face different selection pressures from physiological processes such as micturition, immune response, and antibiotic treatment. Here, we use a mathematical model of the initial phase of infection to unravel the effects of these different selective pressures on the ecological and evolutionary dynamics of UTIs. We explicitly model planktonic bacteria in the bladder lumen, bacteria attached to the bladder wall, and bacteria that have invaded the epithelial cells of the bladder. We find that the presence of non-planktonic bacteria substantially increases the risk of infection establishment and affects evolutionary trajectories leading to resistance during antibiotic treatment. We also show that competitive inoculation with a fast-growing non-pathogenic strain can reduce the pathogen load and increase the efficacy of an antibiotic, but only if the antibiotic is used in moderation. Our study shows that including different compartments is essential to create more realistic models of UTIs, which may help guide new treatment strategies.
Acute lymphoblastic leukemia (ALL) is a type of cancer in which the bone marrow produces too many lymphocytes without further differentiation into mature blood cells. The primary treatment for most ALL cases is chemotherapy; however, after the intense treatment phase, relapse is observed in some patients. Early-stage relapses might be due to some remaining leukemia cells not being detectable by conventional cytomorphology. The molecular tests on minimal residual disease (MRD) can quantify disease burden at relatively low levels, thus helping track cancer remission and relapse. We investigated the dynamics of residual disease to describe the observed response and relapse kinetics heterogeneity. We developed stochastic, patient-specific models based on longitudinal MRD data to further the understanding of the driving factors of relapse, aiming to quantify better prediction of likelihood and timing of relapse in individual patients based on individual treatment choices.
Beta-lactams are the most commonly prescribed antibiotics, and beta-lactamases are ancient enzymes that have evolved to degrade these drugs. When beta-lactamase producing bacteria are exposed to lethal drug concentrations, lysing cells release the enzyme to the medium, thereby contributing to the rescue and eventual recovery of the population. The talk will report on experiments with E. coli exposed to cefotaxime and explain how the observed complex response to near-lethal drug concentrations emerges from a combination of filament formation with the extracellular (public) and intracellular (private) degradation of the antibiotic. Using a minimal mathematical model, it will be shown that, paradoxically, the time to collective recovery is often reduced by increasing the individual death rate.
Extrachromosomal genetic elements play a key role in the evolution of many organisms, particularly in adaptation to environmental stressors such as antibiotics. These elements include plasmids in bacteria, as well as mitochondrial and plastid genomes and extrachromosomal circular DNA (eccDNA) in eukaryotes. They frequently occur in their hosts in multiple copies, allowing increased mutational supply, intracellular genetic diversity, and gene dosage effects. Here we model evolutionary rescue of bacterial and yeast populations by genes of two types of extrachromosomal element—plasmids or eccDNA. Our findings show that bacterial adaptation via plasmids differs from chromosomal adaptation by modulating allele availability and establishment while slowing fixation. Both modelling and empirical validation indicate that segregation can hinder establishment of beneficial plasmid alleles. Furthermore, mutation rates and host population dynamics are influenced by conjugative plasmid transfer and plasmid instability. We further show that plasmid long-term persistence can be secured by a combination of partition systems and toxin-antitoxin systems, which likely play a crucial role in plasmid-mediated evolutionary rescue. For yeast, we develop stochastic models to describe rescue by gene amplification on eccDNA. We show that eccDNA abundance is driven both by intrinsic processes and by the replication of hosts; in particular, biased segregation of eccDNA increases the between-cell variation in copy number. This in turn affects the probability of successful rescue by eccDNA. Overall, we highlight the role of the unique genetic properties of extrachromosomal elements in scenarios of evolutionary rescue, offering insights into how genetic elements persist, adapt, and contribute to
population survival.
Antimicrobial resistance represents a major threat to global health. The spread of resistance is essentially a consequence of evolutionary rescue: the use of antibiotics in medical treatment and animal husbandry can cause dramatic reductions in pathogen population size that may then be countered through evolutionary adaptation in the affected bacteria. Understanding the processes that underlie this undesired form of evolutionary rescue as well as associated trait changes can help us to optimize antibiotic therapy. Here, I will illustrate this rationale with examples from our work with the highly problematic human pathogen Pseudomonas aeruginosa. More specifically, our findings demonstrate that this undesired form of evolutionary rescue can be countered by the sequential administration of antibiotics, especially when: (i) one of the used antibiotics sensitizes the bacterial cells towards another applied antibiotic (i.e., negative hysteresis), (ii) evolution of resistance against one of the antibiotics causes a concomitant increase in susceptibility towards another antibiotic (i.e., collateral sensitivity), and/or (iii) the sequential treatment includes antibiotics with low rates of resistance emergence. A better understanding of the here outlined processes that constrain adaptive responses in pathogenic bacteria may also help us to improve strategies to prevent such undesired evolutionary rescue in parasites, pest species, or cancer cell lines.
In times of accelerating climate change, there is an urgent need for a predictive theory of species’ range shifts, adaptation, and species’ resilience under changing environments. Current predictions are limited, as they rely on theory that neglects important interactions between ecological and evolutionary forces. I will demonstrate that the feedback between selection, genetic drift, and population dynamics fundamentally changes these predictions. This eco-evolutionary feedback creates a tipping point where genetic drift overwhelms selection, and species’ range contracts from the margins or fragments abruptly – even under gradual environmental change. This tipping point is determined by three parameters: two quantifying the effects of spatial and temporal variability on fitness, and one capturing the impact of genetic drift. Importantly, I show that by increasing the ”neighbourhood size” (the population accessible via dispersal), dispersal can counteract the loss of genetic variation, which would otherwise limit evolvability and increase extinction risk. This suggests that increasing local dispersal, possibly via assisted migration, can facilitate evolutionary rescue when local population size is small. While standard ecological niche models assume that species’ ranges shift with environmental change, this study shows that eco-evolutionary feedback can lead to a sudden range fragmentation into disjunct subpopulations with depleted genetic variance, increasing the probability of extinction.
Across the last decade, emergent pathogens have been identified as a common threat to the conservation of charismatic species. Outbreaks of Bd fungus, devil facial tumor disease, and abalone withering syndrome have contributed to considerable population declines of amphibians, Tasmanian devils, and abalone, respectively. In these same populations, there might be evolution of disease resistance and tolerance that can promote evolutionary rescue. However, it is unclear under what general conditions evolutionary rescue might occur in host-pathogen systems. Here, We use a population genetics model to describe evolutionary rescue in host-pathogen systems. We test our model across different dimensions of disease ecology (transmission, compartment structure) and evolution (disease tolerance/resistance/clearance and evolution cost/benefit). We find that evolutionary rescue is more likely to occur in host-pathogen systems when hosts develop strong tolerance (decreased mortality when infected) or clearance (increased speed of recovery). While evolutionary rescue in host-pathogen systems is possible, we also find that evolution of weak or costly disease-robustness can also lead to increased extinction risk. The combined results of our work help conservation managers in identifying populations which are more likely to undergo evolutionary rescue and suggest that conservation supporting population persistence can help promote evolutionary rescue.
Rapid environmental change exposes species to novel conditions, often threatening their long-term survival. Although factors such as genetic variability and mutation rates are well-established drivers of adaptation, the role of predation remains comparatively underexplored. Previous studies suggest that predators can paradoxically facilitate prey adaptation and thereby promoting prey persistence through two mechanisms: the selective removal of less-fit individuals (“selective push”) and the alleviation of density-dependent constraints, which can lead to unexpected increases in prey populations (“hydra effect”). My preliminary simulations, conducted using the NEMO platform as part of my PhD research, support these ideas by indicating that predation enhances prey persistence under changing environmental conditions. Building on these findings, I will adopt a multi-patch framework that incorporates varying degrees of habitat complexity and connectivity, offering insights into how dispersal and environmental heterogeneity shape predator–prey dynamics in a rapidly changing climate.
Indirect evolutionary rescue (IER) is a mechanism where a non-evolving species is saved from extinction in an otherwise lethal environment by evolution in an interacting species. This process has been described in a predator–prey model, where extinction of the predator is prevented by a shift in the frequency of defended towards undefended prey when reduced predator densities lower selection for defended prey. We test here how increased mortality and the initial frequencies of the prey types affect IER. Combining the analysis of model simulations and experiments with rotifers feeding on algae we show IER in the presence of increased predator mortality. We found that IER was dependent on the ability of the prey to evolve as well as on the frequency of the defended prey. High initial frequencies of defended prey resulted in predator extinction despite the possibility for prey evolution, as the increase in undefended prey was delayed too much to allow predator rescue. This frequency dependency for IER was more pronounced for higher predator mortalities. Our findings can help informing the development of conservation and management strategies that consider evolutionary responses in communities to environmental changes.
Almost all organisms have geographical distributions that are limited by range margins. But why? What prevents the evolution of local adaptation in populations at a range margin from allowing the species to expand its range into new territory? One idea is that marginal populations are small and genetically depauperate, with limited potential for local adaptation. Gene flow into such populations would then increase their adaptive potential, effectively providing them the equivalent of evolutionary rescue and enabling further geographic spread. Another idea is that migration into range margins may bring maladaptive alleles compromising the performance of marginal populations and the species’ potential for adaptation and expansion – the opposite of evolutionary rescue. Knowledge on the evolutionary drivers of adaptation across environments not only helps understand range limits but is also key to understand and predict the evolutionary responses of populations facing environmental changes. With EcoRaMa project, we aim to test these two hypotheses using an annual plant, Mercurialis annua, as a model species. M. annua has a distribution ranging from the Mediterranean basins to norther Europe. In 2023, we established experiment populations subject to different scenarios of admixture in 11 gardens across its range and beyond its range margins. We follow the possible evolution of these populations for four years between 2023 and 2027, corresponding to six to eight generations for M. annua. In this poster, we present empirical evidence on climatic adaptation across the range of M. annua in various life history and morphological traits. We also present preliminary results from the first generation showing the importance of germination behavior for influencing demographic performance among populations and experimental gardens.
Multiple populations that are connected by migration create a metapopulation. This study looks at how the fixation probability of a mutant is affected by different migration patterns in various metapopulation structures composed of 4 demes, each having identical carrying capacities. We look at all possible 4-node network structures, where each node is a deme and each link represents migration. We come up with a modified Moran Birth-death process which allows for (at least) partial compensation of change in deme size due to migration. This allows us to analytically explore both symmetric and asymmetric migration in the limit where migration is rare compared to the modified Moran Birth-death process, by adapting the Markov chain method to calculate sojourn times used in Hindersin et al (2014), for metapopulations. We verify the circulation theorem in metapopulations using this method. That is, all metapopulation networks behave the same for symmetric migration. We then compare asymmetric migration towards (and against) more connected demes and its consequences for a mutant in differently structured metapopulations.
New selection pressures can cause rapid evolutionary change. This has, for example, been observed in the Pacific field cricket Teleogryllus oceanicus on Hawai’i, which is under harsh predation pressure from the parasitoid fly Ormia ochracea. The fly locates the male crickets by their mating calls and deposits its larvae on the cricket which gets eventually killed. Within less than 20 generations, a sex-linked mutation causing an alteration to the wing structure became dominant on most islands. Males with this mutation are mute and thus safe from the fly. Yet, mute males cannot attract females from afar. They therefore keep close to the remaining calling males. This mutation creates a trade-off between sexual selection and parasite protection. We use mathematical models to answer under what scenario in this system evolutionary rescue would be possible and how a trade-off causing mutation shapes the dynamics and probability of rescue.
Niche construction, the process by which an organism increases its fitness by modifying its environment, can promote population persistence by allowing niche constructors to restore the density of reproductively suitable habitats. However, niche-constructing populations can be vulnerable to exploitation by non-niche-constructing "cheaters" that benefit from the constructed habitats without paying the cost of production. We first analytically approximated the probability that a declining niche-constructing population undergoes evolutionary rescue by evolving to withstand competition with an invasive cheater. We then evaluated this probability under two different fitness costs of construction: a mortality cost that increases the constructor death rate and a fecundity cost that reduces the constructor birth rate. We find that fecundity costs are less harmful than mortality costs because the former not only reduce the reproductive variance experienced by the mutations responsible for rescue, but also decrease the density of habitats available to cheaters, providing more time for a rescue mutation to appear. We show that the latter benefit can even increase the probability of rescue relative to no cost. Finally, we consider an additional fecundity cost of "niche destruction," whereby an organism reduces its fitness by destroying its own habitat. Such costs are found to promote rescue more effectively than fecundity costs of construction because, in addition to reducing the rate at which new habitats are constructed, niche destruction removes pre-existing habitats that would otherwise be available to invaders. Our analytical approximations are validated with stochastic simulations.
Anthropogenic changes mean that many populations are becoming increasingly small and isolated. The loss of fitness due to inbreeding, aka inbreeding depression, is a major concern for these populations, potentially contributing to their extinction. Genetic diversity can be introduced into the inbred population via the translocation of individuals (Genetic rescue), reducing inbreeding depression. However, the ideal rescuer to use in such situations is highly debated. Rescuers from a large, outbred population which will be more genetically diverse may also risk introducing recessive deleterious alleles. Alternatively, rescuers from other small, inbred populations will likely have purged deleterious alleles but will introduce less new genetic diversity. To explore this issue, we have performed experimental genetic rescue utilising the model species Tribolium castaneum. Inbred populations that have experienced three generations where they were bottlenecked to a single pair, were rescued by either an outbred individual or an inbred individual from another population. Control populations received no rescue. Fitness was measured as the number of offspring surviving to maturity at each generation, over a total of ten generations. After five generations, results show that all rescued populations had increased productivity compared to non-rescued control populations. Those populations rescued by outbred individuals also had increased productivity compared to those rescued by inbred individuals. Our study found that genetic rescue increased the productivity of inbred populations and was more effective utilising outbred rescuers rather than inbred rescuers.
Climate change can increase spatial synchrony of population dynamics, leading to large-scale fluctuation that destabilizes communities. High trophic level species such as parasitoids are disproportionally affected because they depend on unstable resources. Most parasitoid wasps have complementary sex determination, producing sterile males when inbred, which can theoretically lead to population extinction via the diploid male vortex (DMV). We examined this process empirically using a hyperparasitoid population inhabiting a spatially structured host population in a large fragmented landscape. Over four years of high host butterfly metapopulation fluctuation, diploid male production by the wasp increased, and effective population size declined precipitously. Our multitrophic spatially structured model shows that host population fluctuation can cause local extinctions of the hyperparasitoid because of the DMV. However, regionally it persists because spatial structure allows for efficient local genetic rescue via balancing selection for rare alleles carried by immigrants. This is, to our knowledge, the first empirically based study of the possibility of the DMV in a natural host–parasitoid system.
Inbreeding and loss of genetic diversity increase the extinction risk of small isolated populations. Genetic rescue by augmenting gene flow is a powerful means for the restoration of lost genetic variation. In this study, we used multigenerational pedigrees and neutral genetic markers to assess the consequences of outbreeding management in the Chinese and Vietnamese populations of the endangered crocodile lizard, Shinisaurus crocodilurus. Compared with the purebred Chinese population, the outbreeding population exhibited greater molecular genetic variation and a 3-fold larger population size. Moreover, the first-generation hybrids had a longer lifespan than purebreds, suggesting that outbreeding depression did not occur, but the long-term fitness effect of outbreeding needs to be further evaluated. Our study provides valuable insights into the potential for genetic rescue in the endangered crocodile lizard, emphasizing the importance of an evidence-based genetic management approach to address the risks of inbreeding and outbreeding depression in threatened populations.
Global environmental change threatens both the persistence of species and the stability of ecosystems, often demanding rapid evolutionary adaptation. While traditional perspectives have focused on evolutionary change within threatened species, this approach overlooks the crucial influence of evolutionary responses in interacting species. The likelihood and consequences of such indirect evolutionary effects remain poorly understood. In this talk, I discuss the concept of Indirect Evolutionary Facilitation (IEF), a process in which non-evolving populations benefit from the evolutionary adaptations of their interaction partners in response to environmental disturbances. Using a planktonic predator-prey system consisting of six strains of the green alga Chlamydomonas reinhardtii and the rotifer Brachionus calyciflorus, we combine experimental and modeling approaches to investigate how prey evolution affects predator populations under microplastic pollution as a disturbance. Our findings reveal that environmental disturbances can suppress predator growth by limiting ingestion rates, which in turn can drive evolutionary shifts in prey towards increased edibility. Depending on the degree of prey trait variation (as a proxy for evolutionary potential) and the intensity of disturbance, we identify two key processes: Indirect Evolutionary Facilitation, where prey evolution enhances predator density, and Indirect Evolutionary Rescue, where prey evolution prevents predator extinction. These results underscore the pivotal role of indirect eco-evolutionary processes in shaping ecological dynamics. Understanding how trait variation influences ecosystem responses to environmental stressors is essential for predicting species persistence in a rapidly changing world.
Evolution of fungicide resistance in pathogens of crop plants is a prime example of rapid adaptation; and this problem is in many ways similar to evolutionary rescue scenarios. Fungicide resistance causes significant economic losses to crop production that are however difficult to estimate [Mikaberidze et al., 2025]. Fungicides are often applied as mixtures of two components belonging to different modes of action. This exerts a directional selection for pathogen strains resistant to both components of the mixture (double resistance). Since, many important fungal pathogens of crop plants undergo ample sexual reproduction, sexual recombination may accelerate the emergence of double resistance by bringing together mutations conferring resistance to each of the individual components of the mixture. Sex has been previously found to speed up adaptation of yeast populations in vitro [McDonald et al., 2016]. Does a similar effect occur in populations of crop pathogens exposed to fungicide mixtures? To address this question, we formulated a stochastic eco-evolutionary simulation model of pathogen populations that incorporates sexual reproduction and parameterized it for Zymoseptoria tritici, an important fungal pathogen that causes septoria tritici blotch on wheat. According to our model, sex does speed up emergence of double resistance when both single resistant types pre-exist in the population. This happens across wide ranges of strengths of selection and mutation rates. However, when double resistance evolves from a fully sensitive population, sex can speed up emergence of double resistance only over limited ranges of parameters. To gain a deeper insight, we adapted an approximate analytical theory that we devised previously using a combination of a deterministic susceptible-infected-removed model and a stochastic birth-death process [Mikaberidze et al., 2017] to describe these effects and compared the outcomes to full stochastic simulations. We found that in this scenario sex can speed up emergence of double resistance under sufficiently weak selection and for pathogens with sufficiently high basic reproduction numbers; and this parameter range is relevant for typical crop protection scenarios in agroecosystems. Thus, fungicide mixtures alone may prove to be an inappropriate long-term strategy to control crop diseases caused by sexually-reproducing pathogens.
McDonald, M., Rice, D., Desai, M. 2016 Sex speeds adaptation by altering the dynamics of molecular evolution. Nature, 531, 233–236. https://doi.org/10.1038/nature17143
Mikaberidze, A., Gokhale, C.S., Bargues-Ribera, M., Verma P. 2025 The cost of fungicide resistance evolution in multi-field plant epidemics. bioRxiv pre-print https://doi.org/10.1101/2023.09.05.556392
Mikaberidze, A., Paveley, N., Bonhoeffer, S., van den Bosch, F., 2017 Emergence of fungicide resistance: the role of fungicide dose. Phytopathology, 107, 1–16. https://doi.org/10.1094/phyto-08-16-0297-r
The Spotted-Winged Drosophila is an invasive pest that causes millions of dollars of damage to crops in Oregon as well as other places around the world. (Unlike its close relative, the model organism D. melanogaster, this pest feeds on fresh fruit.) While several insecticides are being used to mitigate this damage, there is evidence of decreased efficacy to some due to resistance evolution. In this talk, I will discuss a collaborative project with geneticists and agricultural scientists to identify the genetic basis of resistance and incorporate it into a stage-structured eco-evolutionary model with pulsed selection events. We use this model to identify risk factors and evolutionary rescue patterns during single fruiting seasons as well as long term population trajectories.
Environmental change often occurs abruptly, placing natural populations at risk of extinction unless they adapt rapidly—a process known as evolutionary rescue. While theory and empirical work suggest that genetic variation, population size, and the degree of maladaptation influence the likelihood of rescue, the roles of population history and ecological interactions remain less well understood. This presentation will focus on experimental work carried out on Tribolium castaneum, more commonly known as the red flour beetle. The experiments test how demographic history (e.g. population bottlenecks) and ecological processes (e.g. negative density dependence) influence probability of rescue. The findings of this study demonstrate that both historical and ecological contexts significantly shape evolutionary outcomes. This suggests that these factors must be integrated into predictive models of population persistence. This research highlights controlled experiments based on theoretical expectations, thereby highlighting the value of model-experiment dialogue in refining our understanding of evolutionary rescue. These insights suggest a number of promising avenues for collaboration, particu-
larly in identifying which theoretical predictions merit experimental validation—and vice versa.
Evolutionary rescue helps populations survive environmental change, but the phenotypic and demographic factors associated with rescue dynamics and its long-term effects remain unclear. We experimentally evolved 10 wild-collected populations of flour beetles from across India in a suboptimal corn resource for 70 generations (>5 years), collecting >10,000 population census points book-ended by measurements of fitness-related traits for 30 experimental lines. Despite clear ancestral trait differences, all lines showed highly parallel evolutionary rescue within ~20 generations. Long-term population size varied across source populations and was positively correlated with ancestral development rate, which increased convergently across populations and emerged as the single best predictor of population performance during and after evolutionary rescue. Notably, population dynamics during rescue were uncorrelated both with ancestral trait distributions and post-rescue adaptation. Our results support prior work showing founders as key predictors of adaptation, and highlight the role of founder traits for long-term adaptation and trait evolution following evolutionary rescue.
In addition to its genetic basis, the phenotype on which evolutionary rescue is contingent can depend on non-genetic factors. These exist at different levels of biological organization, including epigenetics (e.g., DNA methylation), cellular and developmental processes (e.g., morphogenesis), behaviour (e.g., cultural traits like tool use), and inter and intra-species interactions (e.g., microbiome). Importantly, these non-genetic mechanisms allow the same genotype to result in multiple phenotypes within the same environment (phenotypic variability) or in different environments (phenotypic plasticity). These changes are not completely dependent on genetic changes and hence can respond to the environmental shift more rapidly. However, the consequences of these rapid changes for survival are non-trivial and depend on the mechanistic underpinning of the non-genetic factor, its interaction with genetic factors, its distinct inheritance patterns, and the timescale of phenotypic changes it causes. Here, I will outline a wide variety of non-genetic factors, discuss existing theory and present two models of evolutionary rescue where the phenotype is composed of a genetic and non-genetic component. The first will focus on a discrete trait with an epigenetic component incorporating multiple epigenetic mechanisms. Increasing epigenetic modification rate has a non-monotonic effect on rescue depending on its interaction with the genetic component and the environment, highlighting the importance of mechanistically modelling such non-genetic factors. The second model will be a general framework for analysing a broad variety of non-genetic factors. Finally, I will discuss the empirical evidence for these non-genetic components in the context of evolutionary rescue highlighting key gaps and open questions.
Evolutionary rescue occurs when one or more populations are declining in abundance and occupancy, and are facing extinction because of environmental change, but adaptation by natural selection occurs sufficiently rapidly to boost mean fitness and abundance, permitting persistence in the changed environments. Understanding when rescue occurs, and when it does not, is pertinent not just to applied questions, such as species persistence in the face of anthropogenic environmental change, and the evolution of resistance to antibiotics and pesticides, but also to a range of classical evolutionary problems. This includes understanding the evolutionary stability of species’ geographical range limits, and broad issues of niche conservatism and adaptive radiations in paleobiology. A concern with temporal variability has been a longstanding focus of research in evolutionary biology, and some classic results suggest that temporal variation in the environment in general hampers adaptive evolution. Temporal variation can arise in selection (e.g., via position of an adaptive optimum, or the width of fitness curves) as well as demographic components of fitness not directly under selection, as well as other factors such as dispersal rates and density dependence. We will touch on all of these. We present theoretical explorations on how temporal variation can be a two-edged sword, sometimes indeed making rescue less likely, but in some circumstances actually enhancing the likelihood of evolutionary rescue and thus population persistence. There is also an intriguing ‘inflationary effect’ of temporal variation on population size that can influence the supply of genetic variation needed to permit rescue. All these recent theoretical insights are ripe for empirical investigation.