Microbial community ecology: bridging theory and observations

11 Jul 2023, 16:00 → 14 Jul 2023, 23:59 Europe/Berlin

Lecture hall - interim building (MPI Plön)

Emil Mallmin (Max Planck Institute for Evolutionary Biology) , Giulia Garcia Lorenzana (LPENS (ENS), MSC (Universite Paris Cite)) , Jules Fraboul (LPENS) , Martina Dal Bello (MIT) , Silvia De Monte (CNRS, IBENS, Paris & MPI Evolutionary Biology, Plon)

Key dates

Workshop dates: July 11 - 14. Starts with a welcome event and dinner from 18:00 on Tuesday 11th. The scientific programme ends 17:00 on Friday 14th.

Registration is closed. All applicants have been informed about the outcome of their application.

 

Overview

Increasingly sophisticated methods are being developed to answer fundamental questions about microbial ecosystems in the wild, in the lab, and in theoretical models:  

• what are adequate descriptors of microbial collectives (genetic, metabolic, abundance-related)?
• what mechanisms enable microbial diversity?
• how do we measure / interpret / model the interactions between microbes?
• how can we predict / design community composition and function?
• what is the role of fluctuations - demographic, environmental, and nonlinear feedback?

To tackle the high complexity of diverse microbial ecosystems, an integration of insights from across different perspectives and methodologies is key. This small-scale workshop aims to create opportunities for fruitful communication across disciplinary boundaries.

The workshop is hosted by the Max Planck Institute for Evolutionary Biology, located in the beautiful lakeside town of Plön (Schleswig-Holstein, Germany).

Organizers: Emil Mallmin, Silvia De Monte, Martina dal Bello, Jules Fraboul, Giulia Lorenzana Garcia

Invited speakers

Ada Altieri, Université Paris Cité

Giulio Biroli, Laboratoire de Physique de l'Ecole Normale Supérieure

Matthieu Barbier, Institut Natura e Teoria Pirenèus (unable to attend)

Elisa Benincà, National Institute for Public Health and the Environment (RVIM) (unable to attend)

Ruben Garrido-Oter, MPI for Plant Breeding Research

Stephan Munch, NOAA Fisheries; University of California Santa Cruz

James O’Dwyer, University of Illinois

Kiran Patil, University of Cambridge

Jarone Pinhassi, Linnæus University

Alexandra Worden, GEOMAR Helmholtz Center; MPI for Evolutionary Biology

Practical information

There is no participation fee.

The workshop will start in the evening of July 11, in order to accommodate long travel schedules.

Travel: directions to MPI Plön. We encourage participants to use train connections whenever this is feasible.

 

The nearest large airport is in Hamburg, from which Plön is reached via regional train or bus in about 2 hours.

Tuesday, 11 July

18:00 → 19:00 Registration & mingle

19:00 → 20:00 Dinner

20:00 → 20:20 Opening statement

20:20 → 21:50 Short introductions: 1/2

Wednesday, 12 July

08:45 → 09:00 Registration (for late arrivals)

09:00 → 09:45 Short introductions: 2/2

09:45 → 10:45 Talks & discussion

Convener: Emil Mallmin (Max Planck Institute for Evolutionary Biology)

09:45 From Patterns to Mechanism to Prediction in Microbiome Science

In this talk we'll explore three different approaches to understanding and prediction in microbiomes. First, we'll explore commonalities between microbial community dynamics and several other complex systems, identifying where general lessons and constraints can inform our interpretation of community patterns. Second, we'll survey a range of mechanistic models, where these have given insights, and also where they fall short in terms of both understanding and prediction. Finally, we'll cover new work on using compressive sensing and machine learning to predict microbiome function from limited data, and whether a deeper understanding of these dynamics can be resurrected from those predictions.

Speaker: James ODwyer (University of Illinois)

10:45 → 11:00 Coffee break

11:00 → 12:00 Talks & discussion

Convener: Jules Fraboul (LPENS)

11:00 Dynamics of high-diversity spatially extended heterogeneous ecosystems

Using statistical physics methods we study the role of dynamical fluctuations in shaping the behavior of high-diversity and spatially extended heterogeneous ecosystems. After an introduction to the methods developed to study the dynamics of species-rich ecosystems, we shall focus on two main emerging phenomena: (i) chaotic dynamics due to endogenous fluctuations persisting for extremely long times, (ii) phase transitions due to demographic noise and dispersal between complete extinction and a self-sustained state with high-diversity.

Speaker: Giulio Biroli (ENS Paris)

12:00 → 13:00 Lunch

13:00 → 14:30 Talks & discussion

Convener: Martina Dal Bello (MIT)

13:00 What I learned from kefir microbiome

Kefir – a traditional fermented milk drink with a long history – is a fascinating example of ecological success at community-scale. The kefir microbial community stands out amongst other fermented food communities with its richness (>40 species), robustness, and spatial organisation. I will present my lab’s findings unravelling the intricate spatio-temporal interactions in the kefir fermentation, discuss implications for understanding principles of community assembly and evolution, and show examples of practical applications. I will also present recent work on gut bacterial communities where similar principles operate and discuss our efforts in modelling community dynamics and evolution.

Speaker: Kiran Raosaheb Patil

13:55 Describing microbial communities using models of cellular resource allocation

To understand the organization and dynamics of microbial communities is a fundamental challenge in current biology. To tackle this challenge, the construction of computational models of interacting microbes is an indispensable tool. There is, however, still a chasm between ecologically motivated descriptions of microbial growth used in typical ecosystems simulations, and the detailed metabolic pathway and genome-based descriptions developed in the context of systems and synthetic biology.

My contribution will outline how models of cellular resource allocation allow us to formulate mechanistic descriptions of microbial growth that are physiologically meaningful while remaining computationally tractable, and therefore offer the potential to advance ecosystem simulations. In particular, recent coarse-grained mechanistic models of microbial growth go beyond Monod-type growth models, and are capable to account for, and explain, several emergent properties of microbial physiology, such as catabolite repression, hierarchies of preferred nutrients, and the utilization of seemingly inefficient pathways.

Of particular interest are trade-offs that arise from limited cellular resources, for example between synthesis of storage compounds versus rapid instantaneous growth. The contribution will discuss the implications of such trade-offs for the organization of microbial community ecology and point out limits in our current understanding of microbial physiology. My main example will be quantitative models of cyanobacterial growth and the emergence of metabolic dependencies between photo- and heterotrophic microorganisms.

Speaker: Ralf Steuer (Intitute for Theoretical Biology, Humboldt-University of Berlin)

14:30 → 14:45 Coffee break

14:45 → 15:45 Talks & discussion

Convener: Giulia Garcia Lorenzana (LPENS (ENS), MSC (Universite Paris Cite))

14:45 KillerProfiles – Unraveling the effects of deeply divergent predators on the evolution of microbial communities

Predation shapes biological evolution at multiple scales, from genomes and organisms to entire ecosystems. Albeit traditionally studied in larger organisms, predation also pervades the microbial world: nematodes and protists ingest prey whole via phagocytosis; many Bdellovibrio-and-like organisms (BALOs) invade their prey and reproduce in a virus-like fashion; and group-hunting myxobacteria deploy a whole arsenal of diffusible and contact-dependent mechanisms to kill and lyse their prey. Despite their expected great impact, little is known about how such deeply divergent predators affect the composition and dynamics of microbial communities over extended evolutionary times. Moreover, it remains unclear how predators adapt to their prey, and whether predator-prey coevolution leads to Red Queen dynamics at the community level. After a short primer on predation in the microbial world, I will present the design of – and ongoing work leading up to – a large-scale evolution experiment that should shed light on these issues. In this experiment, we will use three species each of five divergent classes of predators – myxobacteria, BALOs, dictyostelids, cercozoa, and nematodes – and coevolve each of them with a defined community of 20 soil bacteria. By comparing the effects of each predator species and class on prey community evolution, we will gain systematic insights into how distinct major predators of microbes shape the evolution of species-rich communities. Metagenomic sequencing will moreover allow us to test how different predators and their prey (co-)evolve. Overall, our comparative approach will foster a more comprehensive understanding of the evolutionary ecology of microbial predation.

Speaker: Jos Kramer (Department of Integrative Biology, ETH Zurich)

15:15 The implicit Assumptions of Functional Responses

Research on consumer-resource dynamics is vast and has been addressed in both theoretical and empirical studies. A primary goal in this research agenda is to understand how mascroscopic descriptions of trophic interactions relate to the individual processes that define the consumer-resource interaction in the first place. Despite decades of studies, there is still no clear agreement on the functional forms of the per capita consumption rate of consumers (generally called “functional responses”), not only in community food webs models, but also in simple predator-prey equations. I will present a derivation of functional responses based on the theory of continuous time Markov processes to describe individual feeding interactions, emphasising their stochastic nature. Within this framework, we revisit the derivation of two of the most famous classical functional responses, the Holling Type II FR and the Beddington-DeAngelis FR. We also suggest extensions of them in multi-resource contexts. More importantly, we highlight the common assumptions underlying all these derivations. These implicit assumptions should not be overlooked when using simple models to analyze both experiments and natural consumer-resource interactions.

Speaker: Gian Marco Palamara (University of Bern)

15:45 → 17:00 Optional: Parnass watch tower (2.5km)

15:45 → 17:00 Optional: Walk around lake Schöhsee (5km)

15:45 → 17:00 Optional: swim in the lake

Schösee

17:00 → 19:00 Poster session

17:00 Immunity and competition between bacteria in the gut

Antibiotic resistance is a major threat, motivating the development of procedures to eliminate antibiotic resistant bacteria. Our collaborators (E. Slack’s lab, ETH Zürich) have shown that intestinal antibodies, raised by oral vaccines, enforce the targeted bacterial strain to undergo "enchained growth", forming large clumps, which may be flushed out of the gut faster than free bacteria. They have developed a protocol combining vaccination with the introduction of a niche competitor to eliminate a pathogenic non-Typhoidal Salmonella from the gut lumen. As intestinal colonization is a highly dynamic process, we build a simple mathematical model to generate predictions on the requirements for extinction of the pathogenic strain and the time-to-extinction. We use competition data to estimate key kinetic parameters such as growth rates of the pathogenic strain and the competitor, their loss rates, as well as the carrying capacity. We then estimate the extinction probability and the extinction time distribution for different colonization strategies. Our model confirms that the preventive introduction of the competitor significantly reduces the extinction time of the pathogenic strain.

Speaker: Aurore Woller

17:20 Chaos and noise: disorder in population dynamics

Evolutionary game dynamics is a framework used to model the evolution of strategies in a population. For finite populations, dynamics that appear to lack a pattern or principle of organisation can arise from sources like demographic noise and chaos. The former refers to the stochasticity caused by the probabilistic nature of birth and death events, while the latter is related to deterministic dynamical complexity.
Currently, the effect of noise on dynamics displaying complex behaviour is not yet understood. Therefore, we analyse the interplay between complexity, coming from chaos, and stochasticity, coming from demographic noise. For this, we compare the dynamics that arises from a chaotic deterministic system with the dynamics of the system when demographic noise is added.
In particular, our analysis focuses on quantifying the dynamics’ relevant characteristics via numerical measures. For example, we use tools from evolutionary game theory and chaos theory to quantify the fixation time and the fractal dimension of the system.
Our results confirm the intuitive idea that, for small population sizes, the system’s stochasticity dominates the dynamics. On the other hand, for large enough population sizes, the population dynamics is strongly influenced by the so-called underlying deterministic skeleton, which can exhibit chaotic behaviour.
Overall, our results can help to understand dynamically complex systems affected by demographic noise. Precisely, we found that focusing on the deterministic skeleton can be beneficial to describe and predict complex dynamics of a large but finite population.

Speaker: Alejandra Ramirez

17:40 Genome-assisted characterization and protective role of synthetic microbial community in Arabidopsis thaliana phyllosphere

Plants are associated with a diverse microbiome consisting of bacteria, fungi, and protists that play a crucial role in establishing a stable microbial community that contributes to the health of their host under atypical environmental stresses. Although these microbial communities are co-evolved with plants in either beneficial, commensal, or pathogenic lifestyles, how each member cooperates and contributes to shaping active microbial communities is not clear. In our study, we aimed to understand the role of the stable core leaf microbial community. We characterized culturable representatives' microbes, including bacteria and yeast, to design a complex Synthetic Community (SynCom), resolve microbe-microbe interactions, and investigate their protective outcome under pathogen perturbation. To study their beneficial traits as leaf microbes, we sequenced their genomes and performed in-silico mining of genes/gene clusters responsible for leaf adaptive, protective, and probiotic traits. Genes encoding traits such as colonization, biofilm formation, production of phytohormones, antimicrobials, lytic enzymes, and diverse secondary metabolites were distributed in SynCom microbes with differential abundance, providing insights into their nature of the role and benefit in the microbial community and to the plant. Further, we delved deep into SynCom members and identify key microbe that has a fitness advantage and a protective role under in-vitro and in-planta conditions. We identified biofilm genes/gene cluster abundance and in-vitro biofilm formation ability as an essential factor favoring the functionally effective role of key microbe in the community. Insights gained from this exploration will be useful in designing functionally efficient SynComs for plant protection.

Speaker: Vasvi Chaudhry (University of Tübingen)

18:00 Changes in the root microbiota during the colonization of a pioneer plant after Tropical Glacier retreat

Glacial retreats represent a unique opportunity to study primary successional processes. The new exposed rock is a new material for the assembly of a new ecosystem. Soil formation after glacial retreat is strongly influenced by plant colonization, but this colonization depends on nutrient availability. An ongoing project in the forefront of the last Venezuelan glacier has established a chronosequence of four sites where the glacier retreated between 1910 and 2009. Recent results show a slow successional response during early seral stages. Soil microbial communities play then the key role on ecosystem development through chemical and physical soil modifications that allow subsequent plant colonization. Exploring the mechanisms that assist the establishment of a pioneer plant in these extreme conditions will further our understanding of how the microbiota influences colonization and local adaptation in a novel poor habitat created by glacial retreat. We characterized community structure and diversity of the root endospheres and rhizospheres communities of the pioneer vascular plant in two stages of primary succession of this system (22 and 68 years of ecosystem development). Interestingly, we found that primary succession in belowground communities, as shown for aboveground communities, is also slow. Root-associated bacterial communities are very similar regardless of transect age, suggesting that the environment may be driving these communities, for instance limited initial colonization or severe environmental filtering. Alternatively, community-function in the host-microbiome relationship might be constraining taxonomic diversity. These results highlight the crucial role of host-microbiota mutualisms in the colonization by pioneer species of regions with extreme and quickly changing environmental conditions.

Speaker: Gisela Rodriguez (Microbial Population Biology Department)

18:20 Are diverse soil bacterial communities more resistant to extreme temperature events?

Soil bacteria are critical for sustaining ecosystem functions and services. Recent studies show that soil bacterial communities are susceptible to climate change, particularly to extreme climatic events. Yet, we know little about the biotic mechanisms through which extreme climatic events, such as heat waves, restructure soil bacterial communities. Previous studies indicate that slower growing bacterial species tend to dominate microbial communities as temperatures rise but most of these studies focus on temperatures comfortably within the organisms' thermal niche, assuming a constant community-wide mortality at all temperatures. We will address this research gap using four strains of soil-derived Pseudomonas bacteria, with two slower and two faster growing strains. After measuring the net growth rate and mortality of each strain in monoculture at different temperatures, this thermal niche experimental data will be used to parameterize a null model (i.e., without any biotic interactions) of community growth under various heat shocks and alternative models with only pairwise biotic interactions, or with pairwise and higher-order biotic interactions. To test our theoretical predictions, we will experimentally assemble all combinations of two, three, and four strain communities and measure their resistance to (i.e., immediate response to) and recovery from (i.e., response for several generations after) temperature extreme events. In this way, our work will use communities along a gradient of higher-order interactions to understand how various kinds of species interactions impact community stability under extreme temperature events.

Speaker: Ana-Hermina Ghenu (Institute for Ecology and Evolution, University of Bern)

18:40 Defining features of distributed metabolism in microbial communities

Microbes live in complex communities, where they continuously engage in a range of interactions with other microbes. One class of interactions is the exchange of metabolites, where one microbe synthesises a compound that is taken up by a neighboring microbe. This exchange allows some microbes to be non-producers for essential metabolites, instead relying on the biosynthetic activities of neighbors to obtain nutrients. The metabolic functions of a microbial community can therefore be distributed across individual microbes, leading to a collective metabolism. While it is known that microbial communities from many natural environments exhibit distributed metabolism, the structure and features of these distributed metabolic networks are not well understood. To gain insights into the relationships between microbes in a community, we describe an in silico approach to study how bacteria in a community interact via amino acid exchange, which represents one aspect of a community’s collective metabolism. Our approach uses existing genomic data from natural communities across various environments to construct genome scale models. Based on these models, we will predict how amino acid biosynthetic potential is distributed across communities. Using this data, we aim to understand how environmental variables shape distributed metabolism, and identify any universal features of distributed metabolism that appear across environments.

Speaker: Alyssa Henderson

18:40 Interplay of structure and randomness in complex ecological communities

In recent years, great progress has been made towards understanding the properties of species-rich communities with random interactions by leveraging tools from disordered systems. In the simplest models, macroscopic simplicity comes at the expense of a lack of structure in the interaction matrix. We extend previous work to account for communities with structured weak interactions and explore different types of structure. We characterize a broad class of interaction matrices that give rise to the same phase diagram and equilibrium properties as the unstructured case. Finally, we describe a method to partition communities into a small set of clusters driving the dynamics of species. These clusters can be seen as grouping species having similar "ecosystemic" roles, and provide an intermediate level of coarse-graining in which species are still tracked individually but their dynamics are driven by only a few quantities with a simple ecological interpretation.

Speaker: Juan Giral

Poster.pdf

18:40 Multilevel selection models for microbiomes

Multilevel selection in host-associated microbiomes has important implications for understanding the origin and evolution of these complex associations. To date, we do not have a clear understanding of the different levels that can affect selection on microbial communities. There is evidence that the higher level of selection provided to the microbiome by the host has a significant impact on the evolution of microbial lineages and enhances beneficial interactions [1, 2]. At the individual level, microbial lineages within each host experience selection favoring individuals with higher reproduction rates. Conversely, hosts may have their health status and reproductive success enhanced depending on their microbiome, which, in turn, increases the likelihood of proliferation of the microbiome individuals. For example, let's consider two types of individuals: cooperators and defectors. At the individual level, defectors dominate over cooperators, while at the group level, purely cooperator groups have an advantage over defector groups. Thus, the direction of selection at both levels is in opposite directions - for instance, cooperation is beneficial at a higher level but detrimental for an individual at a lower level.

Although previous studies [1, 2] have investigated the role of a higher level of selection, they focus on the evolution of interactions between a few microbial types. When considering several microbial types, how does multilevel selection influence the selection of interactions? This study aims to understand the possible role of multilevel selection in shaping host-microbiome interactions and how the population structure influences the selection of interactions when there are multiple types of individuals. Does it promote higher microbial diversity in the population? Are microbial lineages more likely to evolve beneficial interactions with their host and other microbes when subjected to multilevel selection? Here, we address these questions structuring a population of individuals into groups. Individuals interact with other group members through an evolutionary game that determines their fitness. Individuals reproduce according to a fitness-dependent Moran process. When the group reaches a certain size, it can split into two groups, while another group is eliminated, or an individual from the group will be eliminated at random. Multilevel selection emerges as a consequence of individual reproduction and constraint from the population structure.

[1] Traulsen, Nowak, PNAS 103, 10952-10955 (2006)
[2] van Vliet, Doebeli, PNAS 116, 20591-20597 (2019)

Speaker: Amanda de Azevedo Lopes (MPI for Evolutionary Biology)

18:40 Multistability of the gut microbiome in response to drug perturbation

The gut microbiota is now well established as a mediator of human health and disease with a central role in a myriad of host functions beyond digestion, including immune modulation, metabolic regulation and neurological signalling. Compositional changes in these bacterial communities have been causally linked to a multitude of diseases including obesity, inflammatory bowel disease, cancer, type 2 diabetes and neurological disorders. However, the dynamics behind these changes and the cause of inter-individual variation in community composition remains poorly understood. A deeper understanding of the response of these communities to perturbations - as potential mediators/triggers of compositional change - is therefore a critical component of establishing the gut microbiota as a promising preventive and therapeutic target in human health. This research aims to combine modelling and experimental approaches to explore the potential multistability of synthetic microbial communities, in particular whether perturbation by antibiotic and non-antibiotic drugs is sufficient to switch between stable states. If this is the case, there is potential to discover hysteric effects in these state switches and further disentangle the complexity of mechanisms in microbial communities.

Speaker: Kate Williamson (MRC Toxicology Unit)

18:40 Parvilucifera s. and microorganism billiards: breaking free through narrow escapes

The microbial ecosystem is full of narrow constrictions that microorganisms need to learn to navigate in order to survive. Here, we study a Nature example of a "microorganism billiard": a system composed of a population of microorganisms packed in a closed space, with only a few narrow apertures to escape from. This situation occurs when the marine parasite Parvilucifera sinerae infects and replicates inside a dinoflagellate host, and the newly born parasites find themselves in the closed and extremely packed space represented by the dead host body (the "sporangium"). In order to start a new successful infection cycle, the parasite's zoospores must find their way out of this closed structure. Which strategies are deployed by the parasites to manage a successful escape? A particular interaction with the boundaries might help them navigate this structure, and collective behaviours between individual parasites might be key in "finding the way out". To understand this phenomenon, we borrow the mathematical formulation of the "active billards", a biological reformulation of the "billard theory" first developed in dynamical systems, and try to extract information about what is happening inside the sporangium from the zoospores escape dynamics. We present the preliminary results of our experiments aimed at characterizing the exit dynamics using high-magnification cameras. Additionally, we introduce the design of a microfluidic chip that facilitates the study of chemoresponses of swimming zoospores to chemical cues.

Speaker: medea zanoli (Mediterranean Institute for Advanced Studies (IMEDEA) , Esporles, Spain)

19:00 → 20:00 Dinner (institute)

20:00 → 21:00 Talks & discussion

Convener: Silvia De Monte (CNRS, IBENS, Paris & MPI Evolutionary Biology, Plon)

20:00 Statistical analysis of ocean microbes

Oceanographers can now collect flow cytometry data in real time while onboard a moving ship, which provides them with fine-scale information about the distribution of phytoplankton across thousands of kilometers. This presents an exciting opportunity to learn new insights about the microbial ecology of the ocean. We present a set of novel statistical models to estimate the time-varying phytoplankton subpopulations while simultaneously identifying the specific environmental drivers. We apply this model to data from numerous oceanographic ships deployed to the North Pacific ocean, and learn new insights about the relationship between microbial populations in the ocean and environmental factors.

Speaker: Sangwon Hyun (University of California, Santa Cruz)

21:00 → 22:30 Drinks & mingle (optional)

Thursday, 13 July

09:00 → 10:30 Talks & discussion

Convener: Giulia Garcia Lorenzana (LPENS (ENS), MSC (Universite Paris Cite))

09:00 Generalized Random Lotka-Volterra model: number of equilibria, glassiness, and beyond

In this talk, I will address some timely questions in theoretical ecology by discussing the Generalized Lotka-Volterra model in the presence of many randomly interacting species and finite demographic fluctuations. Leveraging on disordered systems’ techniques, I will unveil a rich, eventually hierarchical, organization of the equilibria and relate the slowing down of dynamical correlators to glassy-like features. Such a framework works surprisingly well also for integrating and interpreting datasets of patients suffering from Chron's disease, for which a statistically rigorous description is still missing.
Finally, I will discuss possible generalizations to non-logistic growth functions allowing us, on the one hand, to capture positive feedback mechanisms ("Allee effect"), and, on the other hand, to pinpoint new phase transitions as a smoking-gun signature of criticality.

Speaker: Ada Altieri (Université Paris Cité)

09:30 Breakdown of Random-Matrix Universality in Persistent Lotka-Volterra Communities

The eigenvalue spectrum of a random matrix often only depends on the first and second moments of its elements, but not on the specific distribution from which they are drawn. The validity of this universality principle is often assumed without proof in applications. In this talk, I will discuss a pertinent counterexample in the context of the generalized Lotka-Volterra equations. Using dynamic mean-field theory, I derive the statistics of the interactions between species in a persistent ecological community. I will then show that the full statistics of these interactions, beyond those of a Gaussian ensemble, are required to correctly predict the eigenvalue spectrum and therefore stability. Consequently, we see that the universality principle fails in this system. I thus argue that the eigenvalue spectra of random matrices can be used to deduce the stability of “feasible” ecological communities, but only if the emergent non-Gaussian statistics of the interactions between species are taken into account.

Speaker: Joseph Baron (LPENS)

10:00 Local sign stability and its implications for spectra of sparse random graphs and stability of ecosystems

In the very recent years, an increasing amount of interest has been devoted to the study of models of ecosystems defined on sparse random graphs.
In this scenario both network topology and interactions nature play a relevant role in establishing the ecosystem properties.
In particular, differently from what usually happens in dense random matrices, the spectra of locally tree-like graphs with purely predator-prey interactions remain confined in a bounded region on the real axis instead of growing with the system size.
Accordingly, these sparse models does not exhibit a trade-off between size and stability. This feature can provide insights into the modelling of ecosystems, especially with regard to the so-called complexity-stability debate.

Speaker: Pietro Valigi (Università di Roma La Sapienza)

10:30 → 11:00 Coffee break

11:00 → 12:00 Talks & discussion

Convener: Giulia Garcia Lorenzana (LPENS (ENS), MSC (Universite Paris Cite))

11:00 Toward predicting and controlling microbial community dynamics

This talk aims at opening a debate on i) how to address important challenges in predicting microbial community dynamics and ii) how to leverage state-of-the-art ecology to control microbial community dynamics.

I will begin with a brief overview of the dynamics of natural communities in terms of stability in composition and functions, fluctuations, and response to perturbations. Next, I will review recent contributions to understanding community dynamics from statistical physics, systems biology, and experimental approaches.

Then I will highlight some remaining gaps between these fields and motivate a few questions for discussions. How do dynamical regimes (e.g. competition-based vs stochasticity-driven) affect the evolution of community members? How do we move from a statistical understanding to predicting, perhaps controlling, the dynamics of specific communities? How can we leverage transitions between dynamical regimes for specific applications--such as fighting the spread of antibiotic resistance?

Speaker: Daniel Amor (École Normale Supérieure)

12:00 → 13:00 Lunch

13:00 → 14:00 Talks & discussion

Convener: Jules Fraboul (LPENS)

13:00 Single Biotic Interactions Shape Composition of Microbial Communities but Historical Contingency and Evolution Determine Functionality

A major aspiration is to understand the forces that drive composition and functionality of bacterial communities. The abiotic and biotic environment, historical contingency, chance and evolutionary process are all discussed as important factors. In communities, biotic interactions may affect not only composition and functionality but also how species evolve. In the first part I will show that community context can dramatically alter evolutionary dynamics using a novel approach that "cages" individual focal strains within complex communities. We find that evolution of focal bacterial strains depends on properties both of the focal strain and of the surrounding community. In the second part of the talk, I will present latest, yet unpublished insight, that high parallelism in the assembly process is determined by the abiotic environment, but also that individual species have the power to amend composition. However, functionality of the communities depends on historical contingency and evolutionary processes. Finally, I find that, over evolutionary time-scales, interspecific interactions are mainly competitive or neutral in complex communities. The findings demonstrate that adaptation, biotic interactions and historical contingency act in concert to determine community fate.

Speaker: Thomas Scheuerl

13:30 Improving Collective Function in Microbial Communities through Bias-Driven Selection

Collectives of microbes exhibit functions that individual species cannot, such as degrading waste, producing vitamins, and creating biofuels, which can benefit humans. To improve these functions, researchers suggest using artificial selection on collectives to choose the best-performing ones for the next generation. However, this method has shown that there is a limitation to improving the function. In our study, we propose an alternative approach where we select collectives with a bias to counteract natural selection during maturation instead of selecting the best-performing collectives. Our results demonstrate that this strategy leads to further enhancements in collective function by exploring pathways to higher-functioning collectives. Our findings suggest that incorporating a bias will be a promising strategy for improving collective function.

Speaker: Hyejin Park (Department of Physics, Inha University; Asia Pacific Center for Theoretical Physics (APCTP))

14:00 → 14:10 Coffee break

14:10 → 15:10 Talks & discussion

Convener: Silvia De Monte (CNRS, IBENS, Paris & MPI Evolutionary Biology, Plon)

14:10 Artificial evolution of artificial plant- and algae-associated bacterial communities

Photosynthetic organisms, such as land plants and algae, release organic compounds to the surrounding environment, which create a niche for colonization by heterotrophic microbes. These microbes consume photosynthates and assemble into complex communities, often providing their host with beneficial services in exchange, such as pathogen protection, or enhanced nutrient mobilization. While these interactions are ubiquitous in nature, and of great ecological and agricultural importance, direct evidence for adaptation to their host, and simultaneously to other community members remains scarce. Here, we introduce two long-term experimental evolution experiments using artificial microbial communities and multiple algae and plant species, including the model embryophytes Arabidopsis thaliana and Lotus japonicus, and the chlorophyte Chlamydomonas reinhardtii. These reductionist experimental systems allow us to study the dynamic behaviour and evolution of host-associated microbes in a community context and provide data that suggest rapid and reproducible bacterial adaptation to their associated photosynthetic host species and to other microbiota members.

Speaker: Ruben Garrido-Oter (Max Planck Institute / Earlham Institute)

15:10 → 15:30 Coffee break

15:30 → 19:00 Talks & discussion

Convener: Martina Dal Bello (MIT)

15:30 Exploring the effects of metabolic mutations on cell-cell interactions in spatially structured synthetic microbial communities

Microbes are ubiquitous and play key roles in ecosystem functions. In nature, most microbes live in spatially structured multi-species communities. In such communities, cells live in close proximity and engage in a variety of metabolic interactions with neighbours. The overall metabolic potential of a microbial community is determined by the activities of individual cells and the interactions between cells in the local microenvironment. Given their large population sizes and short generation time, microbial communities rapidly mutate and evolve to adapt to the local microenvironment. During this process, a variety of mutants arise. When a mutation that alters a cell’s metabolic activity arises, it may change the interactions the cell has with its neighbours in the microenvironment. These influences can in turn feed back to indirectly change growth and metabolism of the mutated cell. Therefore, newly arising mutations can modify the local metabolic interactions in the microenvironment. The growth of the mutants relative to its neighbours will determine whether or not they will increase in frequency. At present, we do not have a good experimental framework to study what happens when a mutant with altered metabolism emerges in a resident population of spatially organised microbes and to measure the growth and survival of the mutated individual as a function of its cellular neighbourhood.

I use a combination of bottom up quantitative experimental and modelling approaches to study this. In my presentation, I will discuss the experimental set up, the key questions I am interested in testing and the approaches I use to study these questions.

Speaker: Divvya Ramesh (ETH Zürich and Eawag)

16:00 Sugar-acid dichotomy in the carbon catabolic preferences of heterotrophic bacteria

The enormous diversity of heterotrophic bacteria in the environment begs the question to what degree their metabolic niches can be understood in terms of a small number of simplified metabolic categories. Here, we show that, despite high variability at all levels of taxonomy, the catabolic niches of heterotrophic bacteria can be understood in terms of their preference for either glycolytic (sugars) or gluconeogenic (amino and organic acids) carbon sources. This preference is encoded in the total number of genes found in pathways that feed into the two modes of carbon utilization and predictable with a simple linear model based on gene counts, allowing for coarse-grained descriptions of microbial communities in terms of prevalent modes of carbon catabolism. The sugar-acid preference is also associated with genomic GC content, and thus with the carbon-nitrogen requirements of their encoded proteome. Our work thus reveals that how the evolution of bacterial genomes is structured by fundamental constraints rooted in metabolism.

Speaker: Matti Gralka (VU Amsterdam)

19:00 → 20:00 Dinner (at restaurant)

Provided for all participants.

20:00 → 21:00 QA session

21:00 → 22:00 Return walk

Friday, 14 July

09:00 → 10:30 Talks & discussion

Convener: Emil Mallmin (Max Planck Institute for Evolutionary Biology)

09:00 Bacteria-bacteria interactions between gut-associated Bacteroides strains

The mammalian intestine is a unique ecosystem for thousands of bacterial species and strains. How naturally co-existing bacteria of the microbiota interact with each other is not yet fully understood. Here, we systematically studied over 100 interactions between bacteria of the genus Bacteroides that were isolated from the intestine of healthy mice. We find a vast diversity of interactions ranging from positive to negative. Intraspecific interactions are dominated by mutualistic and parasitic interactions. Interspecific interactions are subject to intraspecific diversity and differ between hosts. These findings on obligate host-associated bacteria (i) identify novel molecular mechanisms by which bacteria affect each other and (ii) demonstrate high strain-level variation of bacteria-bacteria interactions. The results have implications for our basic understanding of the microbiota and for the design of synthetic microbial communities.

Speaker: Hanna Fokt

09:55 Mixed strain pathogen populations accelerate the evolution of antibiotic resistance in patients

Antibiotic resistance poses a global health threat, but the within-host drivers of resistance remain poorly understood. Pathogen populations are often assumed to be clonal within hosts, and resistance is thought to emerge due to selection for de novo variants. Here we show that mixed strain populations are common in the opportunistic pathogen Pseudomonas aeruginosa. Crucially, resistance evolves rapidly in patients colonized by multiple strains through selection for pre-existing resistant strains. In contrast, resistance evolves sporadically in patients colonized by single strains due to selection for novel resistance mutations. However, strong trade-offs between resistance and growth rate occur in mixed strain populations, suggesting that within-host diversity can also drive the loss of resistance in the absence of antibiotic treatment. In summary, we show that the within-host diversity of pathogen populations plays a key role in shaping the emergence of resistance in response to treatment.

Speaker: Rachel Wheatley

10:30 → 11:00 Coffee break

11:00 → 12:00 Talks & discussion

Convener: Emil Mallmin (Max Planck Institute for Evolutionary Biology)

11:00 Empirical modeling of ecological dynamics

There is a clear need for robust tools for prediction and inference of ecological dynamics that do not depend on precisely knowing how ecosystems work. Data-driven methods such as empirical dynamic modeling (EDM) allow us to learn dynamics with minimal assumptions. Here, I will introduce the basic ideas of EDM and then discuss two recently developed approaches to incorporating external driving variables that expand our capacity to predict responses to environmental change. In the first, we constrain EDM by modeling dynamics on a "metabolic time step" using the Metabolic Theory of Ecology. This approach improved forecast accuracy in 18 of 19 empirical ectotherm time series (by 19% on average), with the largest gains in more seasonal environments. In the second, we addressed the need to anticipate qualitative changes in dynamics or ‘regime shifts’ in response to a slowly changing driver. Specifically, by incorporating data on the putative driver, we used EDM to reconstruct bifurcation diagrams and infer tipping points in a suite of simulated systems. We applied this approach to empirical time series to predict transitions to, and away from, chaotic behavior as exogenous forcing is changed in an experimental community.

Speaker: Stephan Munch (UC Santa Cruz)

12:00 → 13:00 Lunch

13:00 → 14:30 Talks & discussion

Convener: Martina Dal Bello (MIT)

13:00 On the structuring of bacterioplankton seasonal succession and interannual variation

Seasonality in the conditions for life is a ubiquitous phenomenon in the grand majority of biomes at the Earth’s surface. In aquatic environments, reasonably predictable temporal dynamics in physicochemical conditions of for example temperature and nutrients along with biotic variables set the stage for successional changes in planktonic food webs. This is frequently first noted as changes in phytoplankton dynamics to which bacteria respond but also interactively influence through a variety of feedback processes. These reciprocal interactions have a major influence on the biogeochemical cycles of carbon and macro- and micro-nutrients. The availability of time-series data from a variety of marine stations in different seas currently casts light on the factors that regulate bacterial community structure and the associated bacterial ecosystem functions contributed by different bacterial taxa. This presentation will provide examples of how molecular biology data on bacterioplankton temporal dynamics from the Baltic Sea and other marine environments can be used to gain novel insights into some of the complex processes that shape the seasonality in bacterial growth and its consequences.

Speaker: Jarone Pinhassi (Linnaeus University)

13:55 Nitrogen fixation increases with diazotroph richness in the global ocean

Marine diazotrophs convert atmospheric nitrogen gas into bioavailable nitrogen that can fuel up to 50% of the primary productivity in oligotrophic subtropical and tropical seas. Despite their importance, little is known about their global biogeography and diversity since global studies have been hampered by scarce data observations. This limitation prevents us from understanding the link between diazotroph richness and ecosystem function, especially rates of nitrogen fixation as this is the main bioavailable nitrogen source in oligotrophic oceans. Here, we analyse the correlation between global diazotroph species richness and nitrogen fixation rates by integrating -omics and microscopy-based observations into species distribution models that were developed to cope with uneven sampling effort and sparse observations. Global diazotroph richness is generally higher in subtropical and tropical regions and declines towards the poles. Our results show that sea surface temperature and nutrient-related environmental parameters rank as the most important predictors on a global scale and biogeographic pattern strongly overlap in tropical regions supporting concepts such as niche complementarity. Additionally, we provide the first global biogeographic pattern of non-cyanobacterial diazotrophs that show increased habitat suitability within upwelling regions when contrasted to cyanobacterial diazotrophs. This result further links to the hypothesized niche of non cyanobacterial diazotrophs to be found in nutrient rich waters containing higher concentrations of particulate organic carbon, where oxygen poor microniches provide a sheltered environment for the oxygen sensitive nitrogenase enzyme. We found a positive relationship between global diazotroph richness and nitrogen fixation rates (R = 0.74, p < 0.001; R = 0.66, p < 0.001) supporting the resource use efficiency hypothesis.

Speaker: Dominic Eriksson (ETH Zürich)

14:30 → 14:45 Coffee break

14:45 → 15:45 Talks & discussion

Convener: Silvia De Monte (CNRS, IBENS, Paris & MPI Evolutionary Biology, Plon)

14:45 Illuminating interactions between microbial cells in the wild

We all know that activities and interactions of microbes in the wild underpin the biogeochemical cycles that support earth’s biomes, and the modern climate that we love so well. As in other ecosystems, in the oceans these activities can be difficult to measure particularly in connection with molecular mechanisms or in a manner that identifies the key members responsible for an activity. Likewise, dissecting cell-to-cell interactions, be they between different bacteria, bacteria and protists, or viruses and their hosts, remains a major challenge – impeding the alacrity with which networks of interaction can be modelled. Here we will discuss new insights into microbial activities and interactions in both the surface ocean and the deep sea. The studies rely on a suite of innovations for experimental work in the ocean and ‘single cell genomics’ wherein an individual microbe is captured with a co-associated biological entity, sequenced, and analyzed. Through these studies we are identifying symbioses, pathogenic interactions, virus-host pairs that remain uncultured, and even mechanisms of export for cells that are too small to sink on their own accord from the surface ocean. Collectively, this research identifies new aspects of ecology that shape our knowledge of linkages within marine food webs and the carbon cycle, as well as evolutionary and genomic aspects of microbes.

Speaker: Alexandra Worden (Max Planck Institute for Evolutionary Biology, Plön, DE; Ocean EcoSystems Biology Unit, GEOMAR Helmholtz Centre for Ocean Research Kiel, Kiel, DE; Marine Biological Laboratory, Woods Hole, USA )

15:45 → 16:15 Closing remarks

16:15 → 18:00 End of workshop - Leisure time

18:00 → 23:59 Barbeque (optional)