Predicting ecological dynamics in fluctuating environments

Tuesday, June 15 at 09:30am (PDT)
Tuesday, June 15 at 05:30pm (BST)
Wednesday, June 16 01:30am (KST)

SMB2021 SMB2021 Follow Monday (Tuesday) during the "MS07" time block.
Note: this minisymposia has multiple sessions. The second session is MS06-EVOP (click here).

Share this


Anna Miller (Department of Integrated Mathematical Oncology, Moffitt Cancer Center, United States), Nancy Huntly (Ecology Center and Department of Biology, Utah State University, United States)


Most living organisms, from bacteria to animals, experience temporal fluctuations in their environment. For example, bacteria may face fluctuating periods of antibiotic exposure, or animals experience variations in temperature and resources. To survive through stressful environments, organisms have evolved strategies such as phenotypic plasticity for predictable fluctuations, or bet-hedging for unpredictable fluctuations. Furthermore, temporal variability in resources or treatment can promote coexistence between species or strains that did not occur in constant environments through mechanisms including the storage effect and relative nonlinearity. Mathematical modeling is a useful framework to study how the rate of fluctuations impacts the dynamics of heterogeneous populations, which is useful for a variety of applications including predicting changes to biodiversity due to climate change, or response of coexisting populations of sensitive and resistant cancer cells to treatment. In this minisymposium, we aim to bring together researchers that use both theoretical and experimental approaches in a variety of ecosystems to share ideas centered around the common theme of species evolution and coexistence in fluctuating environments.

Peter Adler

(Department of Wildland Resources and the Ecology Center, Utah State University, USA)
"Challenges in quantifying fluctuation-dependent coexistence mechanisms in nature"
Although modern coexistence theory is 20 yrs old, empirical tests remain scarce. We review the formidable challenges in conducting invasibility analyses in natural ecosystems that make such tests rare. Theory asks, how quickly would each species in a community increase from low abundance in the presence of competitors near their stochastic equilibrium abundances, and how do various features of the environment or the species themselves affect this invasion growth rate? Answering these questions experimentally requires removing a focal species from a community, allowing the remaining species to approach equilibrium, reintroducing the focal species at low abundance, and then repeating these steps under different experimental treatments and for all species in the community. Logistical problems make this approach impractical for macroscopic species growing in nature. An alternative approach is building a model that captures the essential dynamics of the community, and then simulating invasion experiments using the model. The challenges for this approach include naïve application of statistical conventions that may predetermine results, and uncertainty about whether models fit to observational data can accurately project dynamics outside the range of conditions that were directly observed.

Robin Snyder

(Department of Biology, Case Western Reserve University, USA)
"Quantifying fluctuation-dependent coexistence mechanisms for populations of spatially-structured, discrete individuals"
We traditionally analyze coexistence by asking when each species in a system could invade a community made up of the others. To do this, we assume that the invader is rare enough that it does not compete with itself and yet is common enough that we can ignore demographic stochasticity. Spatially extended systems with discrete individuals cause these assumptions to break down. Local dispersal and competition create clumpy invader distributions, so that invaders are common over the scale with which they interact, yet populations are small within the limited scale of interaction, so that discreteness cannot be ignored. Here we present a simulation-based method for quantifying how much different processes or traits contribute to coexistence in spatially structured community models with discrete individuals. We demonstrate our method using simulations of the lottery model and consider contributions from environmental fluctuations (E), competition fluctuations (C), demographic stochasticity, and their interactions. As the spatial scales of competition and dispersal decrease, invaders become more clustered and invader-invader competition increases. This weakens the positive contribution of Cov(E, C) and strengthens the negative effects of fluctuations in C. The effect of demographic stochasticity is small and the trend with increased invader clustering is not statistically significant.

Virginia Turati

(Department of Integrated Mathematical Oncology, Moffitt Cancer Center, USA)
"An integrated approach to understanding the clonal dynamics of childhood B-cell precursor acute lymphoblastic leukemia during treatment to relapse"
Comparison of intratumor genetic heterogeneity at diagnosis and relapse suggests that chemotherapy induces bottleneck selection of subclonal genotypes. However, evolutionary events after chemotherapy could also explain changes in clonal dominance seen at relapse. We investigated mechanisms of selection in BCP-ALL during induction chemotherapy where maximal cytoreduction occurs. To distinguish stochastic versus deterministic events, individual leukemias were transplanted into xenografts and chemotherapy administered. We subsequently leveraged the Hybrid Automata Library (HAL) to implement a mathematical model and, based on the experimental data, infer the evolutionary trajectories leading from initial treatment response to relapse. Analyses of the immediate post-treatment leukemic residuum at single-cell resolution revealed that chemotherapy has little impact on genetic heterogeneity. Instead, treatment acts on the extensive transcriptional and epigenetic heterogeneity of untreated BCP-ALL, selecting a phenotypically uniform population with hallmark signatures of deep quiescence and primitive developmental stage. The mathematical model further suggests that in those leukemias in which most subclones display similar fitness, subclonal selection happens later and not as a direct result of treatment. Instead, in those rarer leukemias in which genotype and phenotypes broadly related to treatment resistance (i.e., proliferation potential) co-segregate, only a few lineages survive through relapse.

Jeff Maltas

(Cleveland Clinic, USA)
"Reversibility of evolution in tunably correlated environments"
Naturally evolving populations constantly face changing environmental conditions. One interesting question is to explore if adaptations that occur as a result of a new environment can be reversed by returning to the previous environment. Using simulations we quantify the genotypic and phenotypic reversibility of an asexually reproducing population. We show that the interlandscape correlation between landscape pairs can dramatically impact the reversibility of this population. Finally, we show that slowly vs quickly switching between landscapes can significantly impact reversibility.

Hosted by SMB2021 Follow
Virtual conference of the Society for Mathematical Biology, 2021.