Schedule for: 22w5140 - Emergent Collective Behaviors: Integrating Simulation and Experiment

A buffet dinner is served daily between 5:30pm and 7:30pm in Kinnear Center 105, main floor of the Kinnear Building.

If you wish to gather on the day of the arrival, TCPL foyer will be open till 10pm. No BIRS staff present.

Breakfast is served daily between 7 and 9am in the Vistas Dining Room, the top floor of the Sally Borden Building.

A brief introduction to BIRS with important logistical information, technology instruction, and opportunity for participants to ask questions.

We often interrogate collective behaviors in the lab in uniform, two-dimensional environments. However, these behaviors and their coordination mechanisms evolved in the wild in irregular, three-dimensional settings. To understand how these behaviors manifest and are coordinated naturally, we have developed a synthetic soil system permitting the growth and starvation-induced development of the cellular slime mold Dictyostelium discoideum. During starvation, Dictyostelium displays collective signaling waves and motility. These behaviors exist in the wild and we are taking a joint theory-experiment approach to link what we understand about single cells and their behaviors in the lab to collective behaviors observed in complex, natural environments.

Organisms can achieve greater actions as a group than as an individual. Population movement and territoriality in ants, birds, and wolves are examples of collective behaviors. Bacteria can also perform these behaviors. In this seminar, I will discuss how bacteria use a local sense of identity to assemble and move as a community. Our unconventional organism, Proteus mirabilis, lives in human and animal intestines and the environment. These bacteria cause disease after moving to the bladder. My research asks how an organism's identity, communication, and local environment influence collectivity.

Collective behavior operates without central control, using local interactions among participants to allow groups to respond to changing conditions. It is widespread in nature, not only producing the coordinated movement of bird flocks or fish schools, but also regulating activity in natural systems from cells, as in cancer metastasis or embryonic development, to the social groups of many vertebrates. An ecological perspective on collective behavior examines how collective behavior adjusts to changing environments. Ant colonies function collectively, and the enormous diversity of more than 14K species of ants, in different habitats, provides opportunities to look for general ecological patterns. Modeling tools from dynamical systems, control theory and distributed algorithms show how local interactions produce the collective foraging behavior of harvester ants in the desert, and generate the trail networks of turtle ants in the tropical forest. These examples suggests how systems with similar dynamics in their surroundings have evolved to show similar dynamics in their collective behavior.

Aggregations are common in ecological systems at a range of scales and may be driven by the underlying landscape in which animals are living or by interactions among individuals. One mechanism leading to ‘self-organized’ animal aggregations is captured by Hamilton’s “selfish herd” hypothesis, which suggests that aggregations may be driven by an individual’s effort to minimize their risk of predation by surrounding themselves with other animals of the same species. Using data captured by satellite imagery and unmanned aerial vehicles (UAVs), we demonstrate that aggregations observed in penguin colonies stem from the delicate interplay between risk avoidance strategies within the population and the landscape terrain in which these dynamics play out. Unlike other animal groups where dynamics are highly fluid (such as bird flocks and fish schools), penguins are highly faithful to individual nest sites; as a result, penguin colonies can become trapped in suboptimal arrangements in which many penguins are stuck breeding on the colony’s edge where their nests are vulnerable to predation. The resulting spatial dynamics are responsible for a hysteretic response to long-term changes in abundance in which even temporary declines in abundance from one year to the next leads to greater fragmentation of the colony, which precipitates further declines. Remarkably, the spatial signature of this process allows us to differentiate a colony that is increasing from one that is decreasing based on spatial pattern alone, which provides even greater insight about the health of colonies captured by satellite imagery. By linking penguin biology with landscape hydrology, geology, and terrain morphology, this work provides a link between current spatial patterning and past dynamics and provides a stark warning about the possibility of critical collapse in populations of these iconic Antarctic species.

Mycorrhizal fungi form an intimate symbiosis with plants and exchange fungal-foraged nutrients for photosynthetically-derived carbon. Mycorrhizal fungal networks connect multiple fungal and plant partners and allow for the transfer of resources and signals between plants. Individual interactions, including plant-plant and plant-fungi, impact group dynamics and shape emergent group-level properties such as ecosystem resilience and stability. Networks transport nutrients, such as N, along source-sink gradients from nutrient enriched to nutrient deficient plants. However, nutrient enrichment can also lower plant reliance on mycorrhizal fungal and thus decrease mycorrhizal fungal abundance, leading to reduced network connectivity and transfer rates. I combine experimental and simulation-based modeling to examine how nutrient pulses influence fungal-plant trading patterns and network properties in order to better understand how these networks facilitate and structure social interactions between plant partners and how these interactions scales up to influence ecosystem resilience and stability.

Lunch is served daily between 11:30am and 1:30pm in the Vistas Dining Room, the top floor of the Sally Borden Building.

Meet in the PDC front desk for a guided tour of The Banff Centre campus.

A buffet dinner is served daily between 5:30pm and 7:30pm in the Vistas Dining Room, the top floor of the Sally Borden Building.

Self-organization can be found in studying crystal formation, superconductivity, social behaviors, etc. It is a challenging task to understand such a phenomenon from the mathematical point of view. We offer a data-driven knowledge-based learning approach to interpret self organization directly from observation data; moreover, our learning approach can aid in validating and improving the modeling of self-organization. We develop a convergent learning framework to derive physically meaningful dynamical systems to explain the observation of first- and second-order self organized dynamics. Next, we study the steady state properties of our estimators. We extend the learning approach to dynamics constrained on Riemannian manifolds. Having successfully applied our learning method to simulated data sets, we study the effectiveness of our learning method on the NASA JPL's modern Ephemerides. In the end, we discuss our current research on learning interaction variables and kernels from observation, and learning from one single snapshot of observation data.

Planar cell polarity (PCP) – asymmetric localization of proteins at cell-cell interface – is essential for embryonic development and physiological functions. Abnormalities in PCP can lead to neural tube closure defects, misalignment in hair follicles etc. Thus, decoding the mechanism responsible for PCP establishment and maintenance remains an open fundamental question. While various molecules – broadly classified into “global” and “local” modules – have been well investigated, their necessity and sufficiency in explaining PCP and connecting their perturbations and defects in experimentally observed patterns has not been examined. Here, we develop a minimal model that captures the proposed features of these two modules- a global tissue level gradient and local asymmetric distribution of protein complexes. Our model results suggest that while polarity can emerge in absence of a gradient, the gradient can provide the direction of polarity as well as offer robustness for maintenance of PCP in presence of stochastic perturbations. We also recapitulated swirling patterns (seen experimentally) and the features of non-domineering autonomy, using only three free parameters in the model - protein binding rate, concentration of proteins forming heterodimer across cell boundaries and steepness of gradient. Our results explain how self-stabilizing asymmetric localisations in presence of tissue-level gradient can lead to robust PCP patterns in diverse biological systems and reveals the minimal design principles for a polarized system.

Despite significant recent advancements, the complex multi-scale brain tumour invasion patterns in 3D are still poorly understood. A particular role in the invasion patterns of the collective migration of the glioblastoma cells populations is likely played by the distribution of micro-fibres, and to address this aspect, in this talk we present our recent advances in this direction. Specifically, we will focus on our recent 3D multiscale mathematical modelling and computational development that builds on our previously introduced 2D multiscale moving- boundary framework and that is able to address the 3D multiscale tumour dynamics. T1 weighted and DTI scans are used as initial conditions for our model, and to parametrise the diffusion tensor. Numerical results show that including an anisotropic diffusion term may lead in some cases (for specific micro-fibre distributions) to significant changes in tumour morphology, while in other cases, it has no effect. This may be caused by the underlying brain structure and its microscopic fibre representation, which seems to influence cancer-invasion patterns through the underlying cell-adhesion process that overshadows the diffusion process.

We discuss microscopic and continuum cell-cell adhesion models and their derivation based on the underlying microscopic assumptions. We analyse the behavior of these models at the microscopic level based on the concept of H-stability of the interaction potential. We will derive these macroscopic limits via mean-field assumptions. We propose an improvement on these models leading to sharp fronts and intermingling invasion fronts between different cell type populations. The model is based on basic principles of localized repulsion and nonlocal attraction due to adhesion forces at the microscopic level. The new model is able to capture both qualitatively and quantitatively experiments by Katsunuma et al. (2016) [J. Cell Biol. 212(5), pp. 561–575].

Motivated by motion of myxo-cells, we consider several kinetic approaches for modeling motion of self-propelled, interacting rods. We will focus on collisional models of Boltzmann type and discuss the derivation of the governing equations, the range of their validity, and present some analytical and numerical results. We will show that collisional models have natural connection to classical mean-field models of nematic alignment.

We examine a chemorepellant model for several groups. Agents from each group move on a two-dimensional lattice, leaving group-specific markings as they move. Agents then avoid areas with markings from other groups. We show that phase separation can occur for certain parameter values. Using a formal derivation, we derive a corresponding cross-diffusive system. Using this continuum system, we are able to identify the critical parameter value at which the separation occurs.

When placed under starvation conditions, the bacterium M. xanthus initiates an aggregation process to ultimately form colonies called fruiting bodies. After initial aggregates have formed, some will destabilize and disappear, with a greater chance of this occurring for smaller aggregates. By analyzing tracked cell data, we identify a bias in the duration a cell moves before changing its motile state when it is aligned with an aggregate. This bias increases with the size of the aggregate, and for smaller aggregates, the bias decreases or even reverses over time. Incorporating this dependence on aggregate size into an agent-based model of aggregate formation driven by experiment data, we show that this change in bias can destabilize small simulated aggregates.

Investigators modeling collective behavior face a variety of challenges involving data from simulation and/or experiment. These challenges include exploring large, complex data sets to understand and characterize the system, inferring the model parameters that most accurately reflect a given data set, and assessing the goodness-of-fit between experimental data sets and proposed models. Topological data analysis provides a lens through which these challenges may be addressed. In this talk, I introduce the core ideas of topological data analysis for newcomers to the field. I then highlight how these topological techniques can be applied to models arising from the study of groups displaying collective motion, such as bird flocks, fish schools, and insect swarms. The key approach is to characterize a system's dynamics via the time-evolution of topological invariants called Betti numbers, accounting for persistence of topological features across multiple scales.

In this talk I will discuss two experimental testbeds designed to investigate behavioral and cognitive response of humans to collective motion across a range of applications. The testbeds utilize virtual reality to immerse human subjects in realistic environments and thus elicit natural responses. In the first test bed, designed to investigate behavioral contagion in human groups, we combine two established pedestrian motion models to create an interactive crowd of virtual characters that walk and shift their gaze momentarily towards a distant overhead location. Participant head orientation and movement are captured as they interact with the virtual crowd to reveal a quorum-like behavioral response. In the second test bed, designed to investigate cognitively responsive strategies for human robot interaction, we mimic an underwater environment after the Great Lakes with five species of fish whose appearance, locomotion, and behavior are modeled based on videos from the field. Participant brain activity and pupillometry data are recorded synchronously in real time to aid in the development of cognitively responsive swarm-robot control strategies.

A lack of Equity, Diversity and Inclusion has long plagued the exact sciences. In Canada, for instance, women hold close to half the number of tenure track position in all of academia, but in physics and mathematics they are outnumbered some four to one. Similarly, visible minorities are outnumbered four to one across all disciplines. Reliable data on LGBTTQ+ representation are hard to get by, but various surveys undertaken by science and engineering departments indicate that some 10 -- 30\% of students identify with this community. It has also been found that this group of student has a lower retention rate and faces barriers specific to their sexual and or gender identity, leading to an under representation at the PhD and tenure-track level. The situation is similar for disabled people, who are represented on the undergraduate level of STEM disciplines at a rate some three times below their representation in the general population. In this panel discussion, we hope to hear the participant's view on this issue, their personal experience with barriers and think of constructive and actionable discussion points to apply to our professional and personal experience. We will be helped by Chad Topaz, who will present some of the work that the Institute for the Quantitative Study of Inclusion, Diversity, and Equity (QSIDE) is doing to leverage data science and mathematical modelling for improving EDI in STEM.

In this talk, I will discuss some work that uses experimental data from in vitro cytoskeletal systems to identify dynamical models and material properties that characterize the system. Two different techniques, the first using statistical modelling and the second, using symbolic regression will be discussed.

In recent years, biological motile cells like bacteria and microalgae have attracted considerable interest not only among biologists but also in the physics community and related fields. Understanding their motion has immense biological and ecological implications. The possibility to harness their motion to power microdevices is a topic of exceptional importance for modern microtechnology. When the motion of a motile cell is observed closely, it appears erratic, and yet the combination of nonequilibrium forces and surfaces can produce striking examples of organization in microbial systems. While our current understanding is based on very simple environments, it remains elusive how, and, at which length scale self-organization emerges in complex geometries. Combining experiments, analytical and numerical calculations [1,2] we study the motion of motile cells and demonstrate that intricate patterns can be observed even at the level of a single cell exploring an isolated habitat. We can explain how curvature guides the motion of the cell. We theoretically predict a universal relation between probability fluxes and global geometric properties that is directly confirmed by experiments [2]. Our results represent a general description of the structure of such nonequilibrium fluxes down at the single cell level. This might open the possibility of designing devices that are able to guide the motion of such microbial cells. [1] J. Cammann, et al., Proc. Natl. Acad. Sci. 118, e2024752118 (2021). [2] T. Ostapenko, et al, Phys. Rev. Lett. 120, 068002 (2018).

Virtual participants are invited to join us for group photo virtually by turning the cameras on and giving a smile. We will take a screenshot of all in gallery view and merge with in-person photo taken earlier this week.

Breast cancer is the most spread cancer in females, with a high mortality rate primarily due to metastasis to secondary sites in the body. The migration of cancer cells away from the primary tumor is influenced by physical interactions between cancer cells and surrounding epithelium and the extracellular matrix (ECM). Cumulative effects of these interactions arise in the form of physical parameters such as: (1) solid stress accumulated within a tumor, (2) surface tension, and (3) viscoelasticity caused by collective cell migration. It is well known that tumor stroma stiffening and solid stress generated in the core region of tumor spheroid during its growth and interactions with external tissue regulate the cancer spreading. Viscoelasticity accompanied with the tissue surface tension are influenced by: (1) the strength of cell-cell and cell-extracellular matrix (ECM) adhesion contacts, (2) intracellular signaling cascades, (3) the viscoelasticity of ECM, and (3) cell contractility in response to micro-environmental conditions. Despite extensive research devoted to study dynamics of cancer progression, we still do not understand the interplay between established factors, which have a feedback impact on the cell rearrangement occurred on various time scales. In order to clarify this issue, it is necessary to consider and compare the rearrangement of various mono-cultured breast cancer and epithelial model systems under in vitro conditions such as: (1) compaction of cell spheroids, and (2) the fusion of two cell spheroids. I’m focusing here on the multi-scale modelling approaches aimed at reproducing and understanding these biological systems.

Active fluids, such as dense suspensions of bacteria or microtubules and molecular motors, display a fascinating range of dynamical states. Active stresses exerted by the individual agents, along with their hydrodynamic interactions, generically lead to the emergence of mesoscale vortex patterns reminiscent of two-dimensional turbulence. In this presentation, we discuss various aspects of turbulence in polar and nematic active fluids. In the first part of the presentation, we explore under which conditions inertial flows can be excited in active nematic turbulence, which typically features low Reynolds numbers. We compare numerical simulations of active turbulence with and without inertial terms. We find that inertia can trigger large-scale motion even for small microscopic Reynolds numbers if the active forcing is sufficiently large and the Ericksen number is sufficiently small. To explain this, we identify an inverse energy transfer caused by inertial advection, whose impact is small in comparison to active forcing and viscous dissipation but which accumulates over time. We additionally show that surface friction, mimicked by a linear friction term, dissipates the transported energy and suppresses the large-scale motion. We conclude that, without an a priori knowledge of model parameters matching experiments, including inertia and friction may be necessary for consistent modeling of active nematic turbulence. In the second part of the presentation, we discuss how ordered flows emerge in the framework of a minimal continuum model for polar active fluids. In particular, we focus on a novel type of turbulence-driven pattern formation: a self-organized active vortex crystal. Crucially, this state emerges from an extended disordered transient characterized by an inverse energy transfer. Exploring the transition from active turbulence to the vortex crystal state, we find analogies to classical phase transitions. For example, we observe locally ordered crystal domains, which share similarities with magnetic domains in ferromagnetic materials, separated by turbulent boundaries. Our results therefore explore one route to self-organization in active fluids.

Contrasting most known bacterial motility mechanisms, a bacterial sliding motility discovered in at least two Gram-positive bacterial families does not depend on designated motors. Instead, the cells maintain end-to-end connections following cell divisions to form long chains and exploit cell growth and division to push the cells forward. To investigate the dynamics of this motility mechanism, we constructed a mechanical model that depicts the interplay of the forces acting on and between the cells comprising the chain. Due to the exponential growth of individual cells, the tips of the chains can, in principle, accelerate to speeds faster than any known single-cell motility mechanism can achieve. However, analysis of the mechanical model shows that the exponential acceleration comes at the cost of an exponential buildup in mechanical stress in the chain, making overly long chains prone to breakage. Additionally, the mechanical model reveals that the dynamics of the chain expansion hinges on a single non-dimensional parameter. Perturbation analysis of the mechanical model further predicts the critical stress leading to chain breakage and its dependence on the non-dimensional parameter. Finally, we developed a simplistic population expansion model that uses the predicted breaking behavior to estimate the physical limit of chain-mediated population expansion. Predictions from the models provide critical insights into how this motility depends on key physical properties of the cell and the substrate. Overall, our models present a generically applicable theoretical framework for cell chain-mediated bacterial sliding motility and provide guidance for future experimental studies on such motility.

Zebrafish are gaining momentum as the third millennium laboratory species for the investigation of several functional and dysfunctional biological processes in humans, including the fundamental mechanisms modulating emotional patterns, learning processes, and individual and social response to alcohol and drugs of abuse. Robotics offers a powerful range of theoretical and experimental approaches that can advance our understanding of this animal model. In this talk, we report recent advances on the integration of biomimetic robotic fish in hypothesis-driven experiments on zebrafish individual and social behavior. We demonstrate the use of robotic fish as highly-controllable and customizable stimuli for laboratory experiments in behavioral neuroscience, paving the way toward systematic validation of mathematical models of animal behavior.

The long-term goal of this project is to understand fundamental principles of fungal-bacterial interactions through physics-based models of metabolism, protein expression and gene expression and to couple these metabolic models to the models of fungal mycelial growth and bacterial chemotactic motion. The parameters for the ordinary differential equations (ODEs) needed to model the dynamics of metabolism are obtained by exploiting the natural selection principle that organisms that have the highest entropy production rates as a group will outcompete species with lower entropy production rates. Because we are using detailed models, entropy production includes growth, maintenance and catabolism. Regulation of the metabolic activity is carried out by a method that combines reinforcement learning of control with statistical thermodynamics and metabolic control analysis, a branch of control theory [1]. Concurrently, structural models of the fungal mycelial growth are being developed. The mycelial models take in glucose, convert the glucose to cell wall precursors which are actively transported through the fungal hyphae, and produce chemo-attractants which are exported and diffuse away in the external environment. These models are coupled with the models of chemotactic motion of Pseudomonas with each bacterium represented by a subcellular element model to describe its structure and mechanical properties [2]. The bacteria run, flick and then reverse directions in order to navigate toward nutrition sources at fungi. The frequency of reversing the direction of motion is controlled by an internal clock. The bacteria have difficulty moving in solid media such as agar (or dehydrated soil) but the water excreted by the fungi due to metabolic activity provides a highway for the bacteria to move on. Ultimately, these bacteria model will also have detailed sub models of metabolisms which will allow us to understand the costs and benefits of metabolic interactions between fungi and bacteria in detail. 1. S. Britton, M. Alber, and W. R. Cannon [2020], Enzyme activities predicted by metabolite concentrations and solvent capacity in the cell, J R Soc Interface, vol. 17, no. 171, p. 20200656. 2. Morgen E. Anyan, Aboutaleb Amiri, Cameron W. Harvey, Giordano Tierra, Nydia Morales-Soto, Callan M. Driscoll, Mark S. Alber, Joshua D. Shrout [2014], Type IV Pili Interactions Promote Intercellular Association and Moderate Swarming of Pseudomonas aeruginosa, Proc. Natl. Acad. Sci. USA vol. 111, no. 50, 18013-18018 (authors for correspondence: J. Shrout and M. Alber).

Complex systems are present across scales in the natural and social world, and the example that I will focus on today is exploring how brightly colored cells interact to form skin patterns in zebrafish. While wild-type zebrafish have black and gold stripes, mutants and evolutionary relatives feature different patterns. As each fish grows, tens of thousands of pigment cells interact through movement, differentiation, and competition to produce its pattern. With the goal of linking genetic mutations to altered cell behaviors and phenotype, I have been developing agent-based models of this process. However, our models how many parameters and are not analytically tractable using traditional methods, making it challenging to fully understand model behavior. On top of this, comparing in vivo and in silico patterns is often a qualitative process, but our models are stochastic and fish patterns are variable. In this talk, I will discuss several approaches to helping address these challenges and more closely relate simulated and experimental data for zebrafish, including methods from topological data analysis, image processing, building more comprehensive agent-based models, and continuum modeling.

Antibiotic resistance has emerged as a global threat. Many argue that we have reached a post-antibiotic era when simple bacterial infections could result in devastating consequences. To address this crisis, extensive efforts are needed to develop strategies to better use existing antibiotics, in addition to developing new ones. In particular, it is critical to understand how bacterial populations respond to antibiotic treatment and how antibiotic treatment can modulate the generation and spread of antibiotic-resistant bacteria. A major mechanism of the rapid spread of antibiotic resistance is horizontal gene transfer (HGT), especially conjugation. In this talk, I will discuss our recent efforts in integrating modeling and experiments to understand the dynamics of the HGT-mediated spread of antibiotic resistance, as well as potential intervention strategies.

Differentiation within multicellular organisms is a complex process that helps to establish spatial patterning and tissue formation. Often, the differentiation of cells is governed by morphogens and intercellular signaling molecules that help to guide the fate of each cell. Here, we couple a synthetic co-repressive toggle switch to intercellular signaling pathways to create a “quorum-sensing toggle”. Our experimental results suggest that this circuit alters the emergent patterns of differentiation in colonies grown on agar containing an externally supplied morphogen. To understand the observed patterns, we developed a coupled 3D PDE-ODE system that takes into account colony expansion. Our bifurcation analysis and simulation results suggest that degradation, diffusion, and sequestration of the signaling molecules are critical to the observed patterns.

5-day workshop participants are welcome to use BIRS facilities (TCPL ) until 3 pm on Friday, although participants are still required to checkout of the guest rooms by 11AM.