Speaker: Yaroslav Tserkovnyak
Date: Friday, February 21, 2020
Location: Physics and Astronomy Building 4-330 at 3:30pm
Title: The role of rare “jackpot” events in the dynamics of evolution and ecology
Speaker: Oskar Hallatschek (UC Berkeley)
Date: Friday, November 22, 2019
Location: Physics and Astronomy Building 4-330 at 3:30pm
Abstract: Luria and Delbrück discovered that mutations that occur early during a growth process lead to exceptionally large mutant clones. These mutational “jackpot” events are thought to dominate the genetic diversity of growing populations, such as biofilms, solid tumors, developing embryos and even invading species. In my talk, I show that jackpot events can be generated not only when mutations arise early, but through other mechanisms that couple evolutionary dynamics to physical processes, which exacerbates their role in adaptation and disease. I will also consider the impact of recurrent jackpot events, which lead to a bias favoring alleles that happen to be present in the majority of the population. I argue that this peculiar rich-get-richer phenomenon is a general feature of evolution driven by rare events.
Title: Forces and mechanosensing in immune cells
Speaker: Arpita Upadhyaya (University of Maryland)
Date: Friday, November 15, 2019
Location: Physics and Astronomy Building 4-330 at 3:30pm
Abstract: Cells need to sense and adaptively respond to their physical environment in diverse biological contexts such as development, cancer and the immune response. In addition to chemical signals and the genetic blueprint, cellular function and dynamics are modulated by the physical properties of their environment such as stiffness and topography. In order to probe and respond to these environmental attributes, cells exert forces on their surroundings and generate appropriate biochemical and genetic responses. These forces arise from the spatiotemporal organization and dynamics of the cell cytoskeleton, a network of entangled biopolymer filaments that is driven out of thermal equilibrium by enzymes that actively convert chemical energy to mechanical energy. Understanding how cells generate forces and sense the mechanical environment (mechanosensing) is an important challenge with implications for physics and biology. We have investigated the principles of cellular force generation, the statistical properties of these forces, and their role in stiffness and topography sensing by immune and cancer cells. During activation, immune cells interact with structures possessing a diverse range of physical properties and respond to physical cues such as stiffness, topography and ligand mobility. We have used traction force microscopy to measure the forces exerted by T cells during activation on elastic substrates. I will discuss the distinct roles of the actin and microtubule cytoskeleton in the exertion of mechanical stresses that support signaling activation, microcluster assembly and receptor movement in T cells. We found two spatially distinct regimes of force generation, potentially arising from different actin-based structures. Furthermore, T cells are mechanosensitive, as cytoskeletal dynamics, force generation and signaling are modulated by substrate stiffness. Our recent studies have also shown that actin dynamics and signaling in B cells is modulated by subcellular topography of the antigen-presenting surface. Our work highlights the importance of cytoskeletal forces in immune cell receptor activation.
Title: The Artificial Axon
Speaker: Giovanni Zocchi (UCLA)
Date: Friday, November 8, 2019
Location: Physics and Astronomy Building 4-330 at 3:30pm
Abstract: Action potentials are the voltage spikes produced in the neuron. Their generation and regulation has been studied in living cells for close to 100 years. The objective of the Artificial Axon is to take this most interesting of dynamical systems out of the cell and into an artificial setting. Create cell-free action potentials, so to speak.
The system is based on the fundamental biological components – ion channels, lipid bilayers – assembled in vitro. It displays the basic electrophysiology of neurons: threshold response and integrate-and-fire dynamics. I will present the system, its shortcomings, and discuss future directions. Namely, we would like to think of the Artificial Axon as the basic unit of an ionic network.
Title: Steering a bacterial pathogen through the phenotype space of multidrug resistance
Speaker: Kevin Wood (University of Michigan)
Date: Friday, October 25, 2019
Abstract: Antibiotic resistance is a growing public health threat. The emergence of resistance far outpaces the development of new drugs, underscoring the need for new strategies aimed at slowing the resistance threat. In this talk, I'll discuss our group's ongoing work to understand the evolution of drug resistance in E. faecalis, an opportunistic bacterial pathogen, using quantitative experiments and theoretical tools from statistical physics and dynamical systems. By combining laboratory evolution with simple mathematical models, we show that unconventional strategies--including aperiodic drug dosing, spatially distributed selection pressures, or adaptive containment protocols--can significantly slow resistance under a surprisingly wide range of conditions. These approaches exploit common (but sometimes neglected) features of microbial evolution--ranging from drug-drug correlations in resistance profiles to spatial heterogeneity and resource competition--to steer adaptation through potentially high-dimensional phenotype spaces. At the same time, the results paint an increasingly nuanced picture of the spatiotemporal dynamics of microbial populations across multiple length and time scales
Title: Single-molecule Experiments with Topologically Complex DNA
Speaker: Alex Klotz (California State University Long Beach)
Date: Friday, October 18, 2019
Abstract: Polymer physics relates the behavior of macroscopic materials to the dynamics of microscopic chain molecules. Experiments with single molecules can bridge the gap between bulk measurements and the statistical behavior of individual polymers. DNA, removed from its biological role, serves as a model system for studying the physics of single polymers, an area of investigation which in turn promotes the development of new genetic sequencing technologies and a better understanding of genomic organization in the cell. I will discuss my work over the past several years studying the physics of polymers with complex molecular topology. These include knotted DNA molecules, which serve as a model to study polymer entanglement, as well as membrane-like catenated (linked-ring) structures called kinetoplasts which can be described as "molecular chain-mail."
Title: Materials that learn from examples
Speaker: Arvind Murugan (University of Chicago)
Date: Friday, October 11, 2019
Abstract: We usually design materials to target desired behaviors that are defined in a top-down manner. Learning theory offers an alternative when desired behaviors are hard to define but easy to give examples of. We consider materials that change as they physically experiences training examples and then test the material on novel inputs never seen before (generalization). We study the physical requirements for such information processing using theory and experiments in two systems: self-assembling DNA strands that can classify high dimensional patterns in chemical concentrations and supervised training of thin sheets to classify high dimensional force patterns.
Title: Collective computation and regulation in the immune system
Speaker: Andreas Mayer (Princeton University)
Date: Friday, May 24, 2019 at 3:30PM in PAB 4-708
Abstract: To combat infection by diverse and ever-evolving pathogens vertebrates have evolved an intricate defense machinery consisting of a large population of highly specialized cells -- the adaptive immune system. To provide efficient defense this system adapts dynamically to the pathogens it encounters. How does a regulated response to a pathogenic challenge arise from the proliferation of individual cells? And how should a well-adapting immune system best adjust to a changing pathogenic environment? We have recently made progress on both questions: First, we have shown that the regulation of T cell expansion by changing antigen-levels explains phenomenological scaling laws of how expansion depends on precursor number, antigen affinity, and antigen kinetics. Second, we have developed a tractable model of how a population of immune cells can optimally integrate information from pathogen encounters with prior expectations. Our work demonstrates the power of simple physical models to help unravel the regulatory mechanisms that shape immune dynamics and the computational principles they implement.
Title: Emergence of Planar Polarity in Tissues and Its Coupling to Geometry
Speaker: Shahriar Shadkhoo (Kavli Institute for Theoretical Physics)
Date: Friday, May 10, 2019 at 3:30PM in PAB 4-330
Abstract: Morphogenesis is the process of the formation of patterns and structures, that shape the body of an adult organism from a highly symmetric initial state. This transition entails cascades of symmetry breaking events, the blueprint of which is encoded in the gene regulatory networks and signaling pathways. Rotational symmetry breaking plays a pivotal role in providing directional information to a multitude of developmental processes, such as cell divisions and growth. Commonly recognized as a manifestation of the broken symmetry, planar polarity arises as a result of anisotropic distribution of signaling proteins on the cell boundaries, and is believed to require cooperative intra- and intercellular interactions, the underlying mechanisms of which are yet to be elucidated. In this talk, I propose a generalized reaction-diffusion model for the binding/unbinding processes of (trans-)membrane proteins. The competition between the two reactions results in local polarization instabilities. The emergence of quasi-long-range polarity then follows from local alignments and the subsequent coarsening of polarized islands across the system. We find constraints on the intracellular interactions that guarantee the stability of tissue polarity, and illustrate how these constraints enable the cytoplasmic proteins to perform a readout of the tissue geometry, as a possible origin of explicit symmetry breaking. I conclude by demonstrating successful recapitulation of some in vivo phenotypes, which allows us to predict the roles of different PCP components through inferring the corresponding model parameters.
Title: Mechanical interactions reduce the power of natural selection in growing yeast colonies
Speaker: Andrea Giometto (Harvard University)
Date: Friday, May 3, 2019 at 3:30PM in PAB 4-330
Abstract: Microbial populations often assemble in dense populations in which proliferating individuals exert mechanical forces on the nearby cells. Using yeast strains whose doubling times depend differently on temperature, we show that physical interactions among cells affect the competition between different genotypes in growing yeast colonies. Our experiments demonstrate that these physical interactions have two related effects: they cause the prolonged survival of slower-growing strains at the actively-growing frontier of the colony and cause faster-growing strains to increase their frequency more slowly than expected in the absence of mechanical interactions. These effects also promote the survival of slower-growing strains and the maintenance of genetic diversity in colonies grown in time-varying environments. A continuum model inspired by overdamped hydrodynamics reproduces the experiments and predicts that the strength of natural selection depends on the width of the actively-growing layer at the colony frontier. We verify these predictions experimentally. The reduced power of natural selection caused by mechanical interactions may favor the maintenance of drug-resistant cells in microbial populations and could explain the apparent neutrality of inter-clone competition within tumors.
Title: Compressed sensing with nonlinear sensory neurons
Speaker: Yuhai Tu (IBM, T. J. Watson Research Center)
Date: Friday, April 26, 2019 at 3:30PM in PAB 4-330
Abstract: Neural sensory systems are capable of coding sparse mixtures of a few stimuli (e.g., different odorant molecules) in a very high dimensional space (the number of possible odorant molecules is huge) by using a relatively small number of receptor neurons (e.g., olfactory sensory neurons). Compressed sensing (CS) is a powerful algorithm developed in computer science and engineering community to efficiently compress high-dimensional information by exploiting the sparsity of the signal. However, the much-celebrated CS algorithm/theory requires the sensors to be linear. For sensory systems such as the olfactory system, the receptor neurons respond nonlinearly to odorant concentration and have a finite response range. Therefore, the CS algorithm does not apply to sensory systems, and how sensory systems compress information remains an open question.
In this talk, we will present some recent results on how a relatively small number of nonlinear sensors each with a limited response range can optimize transmission of high dimensional sparse odor mixture information. We found that the optimal coding matrix is ``sparse'' -- only a subset of sensors respond to a given odorant and the sensitivities follow a broad (such as log-normal) distribution matching the odor mixture statistics. We showed that this maximum entropy code enhances the performances of the downstream reconstruction and classification tasks. Our study showed that introducing odor-evoked inhibition further enhances coding capacity for neurons with a finite spontaneous activity. Comparisons with available experiments in olfactory systems are consistent with our theory.
Title: Size Coordination in Animals and Statistical Physics of Growing Tissues
Speaker: Ojan Khatib-Damavandi (University of Michigan)
Date: Friday, April 19, 2019 at 3:30PM in PAB 4-330
Abstract: One of the major questions in biology is how organs know when to stop growing and how animals coordinate growth of their organs to ensure proper body proportions. In this talk, we first discuss coordination between left and right organs (like fruit fly wings) via chemical signaling. We show analytically that there are limits to the ability of the signal to ensure successful left/right symmetry, suggesting that organ sizes are primarily set autonomously. Inspired by this conclusion, we look at noisy growth of individual tissues subject to mechanical feedbacks and show that even the simplest model of tissue growth exhibits a surprisingly rich behavior. For instance, we find that the growth displays power law correlations and soft modes that lead to large variations in the size of marked clones of cells.
Title: Cooperative Survival of Bacteria Under Attack by Antimicrobial Peptides
Speaker: Sattar Taheri-Araghi (California State University, Northridge)
Date: Friday, April 12, 2019 at 3:30PM in PAB 4-330
Abstract: Antimicrobial peptides (AMPs) are broad spectrum antibiotics that utilize electrostatics to selectively target bacteria. Like all antibiotics, AMPs need a minimum concentration to inhibit growth of a bacterial culture. Yet, we do not have a clear quantitative picture of the dynamics of their action in a population of bacteria. In this talk, I present our recent finding on how heterogenous absorption of AMPs in dead bacteria allows survival of a bacterial population under attack by AMPs. Our single-cell analysis demonstrate a rapid absorption and retention of a large number of AMPs by Escherichia coli cells upon the inhibition of their growth. Thus, cultures with high cell density exhibit two distinct subpopulations: a non-growing population that absorb peptides and a growing population that survive owing to the sequestration of the AMPs by others. A mathematical model based on this binary picture reproduces some rather surprising observations of population dynamics, including the increase of the minimum inhibitory concentration with cell density (even in dilute cultures) and the extensive lag in growth introduced by sub-lethal dosages of LL37 peptides.
Title: Spontaneous Emergence of Chirality in Achiral Systems
Speaker: Mohan Srinivasarao (Georgia Tech and National Science Foundation)
Date: Friday, April 5, 2019 at 3:30PM in PAB 4-330
Abstract: Lyotropic chromonic liquid crystals (LCLCs) are a relatively new class of liquid crystals (LCs) that have attracted considerable attention in recent years. Applications of these materials have been explored as polarizers, optical compensators, biosensors, precursors of aligned graphene and templates for mesoporous nanofibers. LCLCs consist of many dyes, drugs, nucleic acids, antibiotics, carcinogens and anti-cancer agents. In this talk I will explore the spontaneous emergence of chiral structures from achiral lyotropic chromonic liquid crystals when confined to cylindrical capillaries with various boundary conditions. When confined to a cylindrical geometry with planar boundary conditions, the presumed ground state of a nematic fluid corresponds to that of an axial configuration, where the director, free of deformations, lies along the long axis of the cylinder. However, upon confinement of lyotropic chromonic liquid crystals in cylindrical geometries, we uncover a surprising ground state corresponding to a doubly twisted director configuration. The stability of this ground state, which involves significant director deformations, can be rationalized by the saddle-splay contribution to the free energy. It will be shown that sufficient anisotropy in the elastic constants drives the transition from a deformation-free ground state to a doubly twisted structure, and results in spontaneous reflection symmetry breaking with equal propensity for either handedness. Enabled by the twist angle measurements of the spontaneous twist, we determine the saddle-splay elastic constant for chromonic liquid crystals. I will also discuss the path to a monodomain or a single crystal, if you will, of chrominc liquid crystals confined to a rectangular capillary enabled by a spontaneous twist deformation.
Title: Quantifying landscape and flux for nonequilibirum biological systems
Speaker: Jin Wang (Stony Brook University)
Date: Friday, March 22, 2019 at 3:30PM in PAB 4-330
Abstract: We established a theoretical framework for studying the dynamics and thermodynamics of nonequilibrium physical and biological systems. We identify the main driving force for the nonequilibrium systems as the landscape gradient and rotational flux. We found that landscape and flux can be quantified and are critical for the global dynamics and thermodynamics of the nonequilibrium systems. We uncovered that these driving forces are crucial in determining the functions of cell cycle, differentiation/development, cancer, neural networks, evolution and ecology. We demonstrated that the landscape and flux can be quantified experimentally in self repressor/lambda phage and single molecule enzyme dynamics respectively.
Title: Principles of effective immune responses to viral infections – examples from influenza and HIV
Speaker: Ruian Ke (Los Alamos National Laboratory)
Date: Friday, March 15, 2019 at 3:30PM in PAB 4-330
Abstract: The immune system is a highly orchestrated system that serves as a critical line of defense in response to viral infections. In this talk, I will first present a work on understanding the principles of robust innate immune response to viral infections (such as influenza) immediately after virus exposure. Using a combination of mechanistic models, network models and stochastic cellular automata models, we show that the innate immune response uses cell-to-cell communication to build up a layer of protected cells to effectively constrain infection at the local area of exposure - a strategy reminiscent of ‘ring vaccination’ in disease outbreak control. Interestingly, this strategy seems to be robust against variations in the host cell arrangements, providing an explanation of why this is used as a general host response to a wide range of viral infections. If time permits, I will present our recent work on understanding the immune control of HIV after treatment interruption. We propose a theory that stochasticity in HIV reactivation and infection can lead to failure of the immune control, resulting in HIV rebound. Using the WKB approximation, we identify key determinants of the time to HIV rebound. These results have broad implications for the design of immuno-therapies to boost immune control to achieve life-long HIV remission.
Title: The multi-scale structure of chromatin in the nucleus
Speaker: Yuval Garini (Bar Ilan University)
Date: Friday, March 1, 2019 at 3:30PM in PAB 4-330
Abstract: The DNA in a human cell which is ~3 meters long is packed in a tiny nucleus of ~10 μm radius. The DNA is immersed in a condensed soup of proteins, and it is highly dynamic while taking part in many processes such as protein expression and cell division. Nevertheless, it must stay organized to prevent chromosome entanglement. Studying this nanometer – micrometer scale structure is difficult, as it requires to use both high spatial and temporal resolutions for studying its characteristics. We adopted various methods for studying the organization of the genome in the nucleus, including live-cell imaging, time-resolved spectroscopy, chromosome conformation capture (3C) and single molecule methods such as AFM. These methods are followed by biophysical modeling. It allowed us to identify that a protein, lamin A, forms chromatin loops thereby restricting the chromatin dynamics in the whole nucleus volume. Based on the results, we conclude that the organization of the DNA in the nucleus is based on a “chromatin matrix”, a structure that we describe here for the first time. I will describe the problem, the methods we use, the results and the conclusions as described above.
Title: Topological mechanical edge floppy modes in aperiodic systems: disordered fiber networks and quasicrystals
Speaker: Di Zhou (U of Michigan)
Date: Friday, February 22, 2019 at 3:30PM in PAB 4-330
Abstract: Maxwell lattices are spring-and-mass frames with balanced degrees of freedom and constraints, and are thus on the verge of mechanical instability. They exhibit zero-frequency mechanical edge floppy modes that are robust against disorder, randomness, and stochastic damage. These protected modes are the mechanical analogue of topological states in electronic systems. However, the concept of topological protection is not limited to spatially periodic systems. In this talk, we show how topological edge floppy modes arise in messy systems. We study disordered fiber networks and quasicrystals to demonstrate how these modes are generated through geometric changes in the frames. Fiber networks are not only important in understanding a broad range of natural and manmade materials (such as cytoskeleton and porous media) but also exhibit rich physics. Quasicrystals are fascinating structures with properties not available in crystalline structures, such as rotational symmetry. The study of topological mechanics in messy systems opens the door to new physics in biological networks as well as novel designs of topological mechanical metamaterials.
Title: Phase separation of multicomponent liquid mixtures
Speaker: Andrej Kosmrlj (Princeton)
Date: Friday, February 1, 2019 at 3:30PM in PAB 4-330
Abstract: Multicomponent systems are ubiquitous in nature and industry. While the physics of binary and ternary liquid mixtures is well-understood, the thermodynamic and kinetic properties of N-component mixtures with N>3 have remained relatively unexplored. Inspired by recent examples of intracellular phase separation, we investigate equilibrium phase behavior and morphology of N-component liquid mixtures within the Flory-Huggins theory of regular solutions. In order to determine the number of coexisting phases and their compositions, we developed a new algorithm for constructing complete phase diagrams, based on numerical convexification of the discretized free energy landscape. Together with a Cahn-Hilliard approach for kinetics, we employ this method to study mixtures with N=4 and 5 components. We report on both the coarsening behavior of such systems, as well as the resulting morphologies in 3D. We discuss how the number of coexisting phases and their compositions can be extracted with Principal Component Analysis (PCA) and K-Means clustering algorithms. Finally, we discuss how one can reverse engineer the interaction parameters and volume fractions of components in order to achieve a range of desired packing structures, such as nested "Russian dolls" and encapsulated Janus droplets.
Title: Spatial stochastic simulation for interacting particle systems
Speaker: Tim Stutz and Alfonso Landeros (UCLA)
Date: Friday, January 25, 2019 at 4:00PM in PAB 4-330
Abstract: Dynamics of microenvironmental regulation of tumor growth and invasion are poorly understood. Low counts of key molecular and cellular species introduce stochasticity into niche regulatory networks, and these systems are inherently spatial. We present a simulation approach for modeling spatially homogeneous stochastic systems, by introducing an extension of the n-fold approach to Kinetic Monte Carlo that connects to simulation techniques for well-mixed stochastic processes. Our general purpose approach performs well on interacting particle systems, preserving volume exclusion, and modeling diffusion via particles obeying random walks respecting exclusion. The software provides an intuitive interface to observe behavior of the multiple interacting species within cancer cell microenvironments. Spatial heterogeneity is vital for accurately modeling many other evolutionary processes, including desertification, disease spreading, and maintenance of species biodiversity, and our algorithm can be generalized to each of these processes. We provide examples from ecology and immunology to illustrate the flexibility of our software.
Title: Tackling noisy dynamics in the epigenome with stochastic modeling and inference
Speaker: Elizabeth Read (UCI)
Date: Friday, December 7, 2018 at 4:00PM in PAB 4-330
Abstract: Increasingly, cell biological data is available at single-cell, single-nucleotide, and single-molecule resolution. Such experiments reveal often-unexpected levels of heterogeneity at these scales. In order to quantitatively describe this heterogeneity and accurately infer mechanistic principles from experimental data, stochastic models are often essential. However, a number of challenges currently face stochastic modeling in cell biology, from designing mathematical frameworks that accurately reflect biological complexity, to computation and data-integration.
This talk will present ongoing work in the Read lab in two areas related to epigenetic regulation. First, stochastic models coupled with novel simulation techniques shed light on gene regulatory network dynamics during embryonic development, and begin to reveal general principles of cell-phenotype stability versus plasticity. Second, stochastic-model-aided analysis of experimental bisulfite sequencing data reveals significant temporal heterogeneity in DNA methylation patterns post-replication, suggesting novel mechanisms of dynamic epigenetic regulation.
Title: Space and Time in Genomic Interactions
Speaker: Olga Dudko (UCSD)
Date: Friday, November 30, 2018 at 4:00PM in PAB 4-330
Abstract: Many processes in biology, from antibody production to tissue differentiation, share a common fundamental step — establishing physical contact between distant genomic segments. How fast this step is accomplished sets the “speed limit” for the larger-scale processes this step enables. A key outstanding question is then: How do genomic segments that are strung out over millions of base pairs along the DNA find each other in the crowded cell on a remarkably short timescale? This question, fundamental to biology, can be recognized as the physics problem of the first-passage time. I will show how concepts from statistical physics help reveal the physical principles by which cells solve this first-passage problem with astonishing efficiency. I will illustrate these ideas in the context of V(D)J recombination – the genetic mechanism that allows the human immune system to respond to millions of different antigens.
Title: Emergent eco-evolutionary phenomena in microbial communities
Speaker: Kalin Vetsigian (University of Wisconsin)
Date: Friday, November 9, 2018 at 4:00PM in PAB 4-330
Abstract: The complexity of microbial community dynamics stems not only from the diversity of these communities and the richness of their microbial interactions but also from the fact that many of these interactions can readily evolve. As mutant strains with altered interactions increase in frequency they reshape the ecological dynamics and the selection pressures on existing strains. The spectrum of possible consequences of such an interplay between ecology and evolution are poorly understood. To start filling this gap, we computationally investigated the eco-evolutionary dynamics in communities dominated by toxin-mediated interactions. Such interactions are ubiquitous among soil microbes, and whether and how they contribute to diversity has been a long-standing puzzle. We identified several emergent eco-evolutionary phenomena. First, the dynamics could robustly discover complex evolutionary stable states in which multiple strains coexist (Nash equilibria) despite the fact that such states are unreachable through a step-by-step community assembly. Rather, the system as a whole tunnels between collective states via a fundamentally eco-evolutionary process. Second, communities of particular strains can emerge and persist even if these communities are not ecologically stable. Finally, the dynamics can exhibit intermittency in which prolonged periods of apparent community stability are interrupted by periods of fast strain turnover. In spatially structured communities, this intermittency leads to mosaics in which different spatial regions are in different eco-evolutionary regimes in a phenomenon reminiscent of phase coexistence in material science. These findings demonstrate that toxin-mediated interactions are a viable mechanism for explaining diversity, provide a qualitatively new mechanism for adaptive diversification, and expand our understanding of the different possible modes of eco-evolutionary dynamics in microbial communities.
Title: The conundrum of the nuclear pore: enhanced diffusion by binding to a polymer gel
Speaker: Carl Goodrich (Harvard)
Date: Friday, October 12, 2018 at 4:00PM in PAB 4-330
Abstract: Creating a selective gel that filters particles based on their interactions is a major goal of nanotechnology, with far-reaching implications from drug delivery to controlling assembly pathways. This is particularly difficult when the particles are larger than the gel’s characteristic mesh size because such particles cannot passively pass through the gel, and it is not clear what physical principles a designer gel could exploit to attain such selective permeability. Nature, however, appears to have found a solution in the form of the nuclear pore complex, which is a massive protein structure with a gel-like plug that regulates the transfer of macromolecules between the nucleus and the cytoplasm, but the mechanistic details of the nuclear pore are not known. We study a highly simplified model of a crosslinked polymer gel, and present an equilibrium mechanism where crosslink binding dynamics are affected by interacting particles such that particle diffusion is enhanced. In the limit of high binding affinity, this mechanism leads to a perfect filter where large binding particles diffuse through the gel while non-binding particles are permanently trapped. In addition to revealing specific design rules for manufacturing selective gels, our results have the potential to explain the origin of selective permeability in the nuclear pore complex.
Title: Trying to understand the brain from the bottom-up: topologically well-defined neural networks in-vitro
Speaker: Csaba Forro (ETH Zurich)
Monday, August 13, 2018 at 2:00PM in PAB 4-330
Abstract: In the search of unveiling the fundamental mechanisms of brain function and brain computation, we believe that the right scale to look at is the small network scale. Indeed, elementary and identifiable computations are performed by only a handful of neurons in systems like the C.Elegans or the retina. Current microelectrode technologies paired with in-vitro neuroscience offer an appealing solution to this end. Indeed, advances in microelectrode arrays allow to probe neural activity with a temporal resolution of tens of microseconds but also with a spatial resolution of tens of micrometers. In order to disentangle the immense complexity and connectedness of neural structures in-vivo, in-vitro neuroscience provides tools to build neural networks from the bottom up in a controlled fashion.
To achieve this, we use Polydimethylsiloxane (PDMS) microstructures akin to small labyrinths, in which through-holes allow for cell-body compartmentalization. These compartments are connected with shallow micro-tunnels that are permissive only to axons and neurites. By using special channel geometries, we achieve up to 95% directional axon guidance from compartment to compartment. This was put to use in making clockwise-connected neural networks of 3 nodes, which were compared to their non-directed counterparts, and we found that information flow - in terms of transfer entropy - from node to node was up to 100 times higher in directed networks, highlighting the importance of controlling the connectivity of cultures.
Title: The Role of Sub-cellular mRNA Localization in Bacterial Gene Expression
Speaker: Hamid Teimouri, Harvard University
July 20, 2018 at 4:00pm in PAB 4-330
Abstract: Unlike their eukaryotic counterparts, bacterial cells are composed of a single compartment. This allows many rapidly diffusing macromolecules, such as proteins and mRNAs, to be evenly distributed in the cell. Important exceptions are proteins embedded in the cell membrane, which transport material and information across the membrane. Often these proteins attach to the membrane before their translation is complete, anchoring their mRNAs to the vicinity of the membrane.
This coupling between translation and localization suggests that the dynamics of translation may shape the spatial organization. After reviewing experimental observations, I will discuss our recent work using a canonical model of nonequilibrium statistical physics to characterize this connection and show how tunable kinetic properties allow the cell to regulate the spatial organization of both mRNAs and proteins. Moreover, subcellular distributions of mRNAs raise the question how localization may affect the regulatory activity of small noncoding RNAs. In the second part of my talk, I will discuss our compartmental model that addresses this question. Our results suggest that under certain conditions, interpretation and modeling of natural and synthetic gene regulatory circuits need to take into account the spatial organization of the transcripts of participating genes.
Title: Correlating structure and dynamics in glassy liquids and model tissues
Speaker: Andrea J. Liu, University of Pennsylvania
June 1, 2018 at 1:30pm in Young Hall 2-051 (via Skype)
Abstract: In a liquid, any constituent particle can rearrange and change its neighbors, but in crystalline solids, only particles near structural defects such as dislocations are prone to rearrange. A decades-old question in the field of supercooled liquids has been whether all particles are equally likely to rearrange, and if not, what are the “defects” that make some particles more likely to rearrange than others? I will show that machine learning provides a powerful and efficient approach for identifying local structural environments that are predisposed to rearrange. We use machine learning to define a quantity, “softness,” that is highly predictive of rearrangements and simplifies our understanding of glassy dynamics considerably. Finally, we apply the same analysis to models that quantitatively describe real epithelial tissues and find that softness is quite predictive of cellular rearrangements.
Title: Mechanics of Allosteric Materials — How Proteins Mediate Long-Range Interaction
Speaker: Le Yan (KITP)
May 18, 2018 at 4pm in PAB 4-330
Abstract: A crucial regulation for life is enabled at the molecular level through allosteric proteins, whose catalytic activity at the active sites can be significantly enhanced or inhibited by the appearance of specific chemicals binding at their allosteric sites. Predicting the allosteric pathways from protein structures would help us determine regulation networks and design smart drugs. However, it remains a challenge as the nature of allostery — this elasticity-mediated long-range interaction is understood superficially. To approach the problem with systematic samples, we introduce a numerical scheme with the model proteins evolving in-silico to accomplish specific allosteric functions. We then obtain statistical features among thousands of solutions to the tasks and deduce rules for the allosteric mechanics. For the geometric task when a specific strain is propagated through the media, we find commonly applied is an edge-mode mechanism which quickly amplifies the strain signal close to the active site. While for the cooperative task with elastic energy conveyed, we reveal that the appearance of a mechanism, an extended and nearly zero energy mode, dominates the pathway. This mechanism can appear as shear, hinge, and twist — designs found in some allosteric proteins, or be more complicated and hard to visualize. But independent of specific designs, the cooperative energy stored in the mechanism does not decay with the size of the protein or the distance between the active and allosteric sites.
Title: Topology in Polar Flocking and Active Nematics
Speaker: Mark Bowick, Kavli Institute for Theoretical Physics, UCSB
May 11, 2018 at 4pm in PAB 4-330
Abstract: Active flocking on curved surfaces, such as the 2-sphere and the catenoid, exhibits dynamical symmetry breaking in the form of spontaneous flow, calculable inhomogeneous density patterns and long-wavelength propagating sound modes that get gapped by the curvature of the underlying substrate. Curvature and active flow together result in symmetry-protected topological modes that get localized to special geodesics on the surface. These modes are the analogue of edge states in electronic quantum Hall systems and provide unidirectional channels for information transport in the flock, robust against disorder and backscattering. Active nematics instead exhibit spontaneous motility of strength +1/2 disclinations and active torques that favor the motility-driven unbinding of defects. Despite the directed motion of defects, nematic order is stabilized by rotational noise at low enough activity. Within a perturbative treatment, active forces lower the effective defect-unbinding transition temperature.
Title: Hydrodynamics of transient cell-cell contact in T cell receptor triggering
Speaker: Jun Allard, UC Irvine
May 4, 2018 at 4pm in PAB 4-330
Abstract: In many biological settings, two or more cells come into physical contact to form a cell-cell interface. Examples include wound healing, tissue development, tumor growth, and some bioengineered diagnostic tools. In some cases, the cell-cell contact must be transient, lasting only seconds. One important example is offered by the T Cell, an immune cell which must attach to the surface of other cells in order to decipher information about disease. The aspect ratio of these interfaces (tens of nanometers thick and tens of micrometers in diameter) puts them into the thin-layer limit, or “lubrication limit”, of fluid dynamics. A key question is how the molecules on the two cells (receptors and ligands) come into contact. What are the relative roles of thermal undulations of the plasma membrane and deterministic forces from active filopodia? We present a computational fluid dynamics algorithm capable of simulating fluid-structure interactions with thermal fluctuations on seconds- and microns-scales. We use this to simulate two opposing membranes, variously including thermal fluctuations, active forces, and membrane permeability. Our results demonstrate that the force required will increase for smaller cell-cell distances (where the thin-layer effect is strongest), leading to an optimal “attack range” that might explain the observed geometry of filopodia. The results also suggest an important role for membrane permeability; Factors that influence permeability, such as aquaporins, are dynamically controlled by the cell, and have been shown to impact cell processes including cancer angiogenesis, raising the possibility that cell-cell contacts can be regulated in this way.
Title: Soft matter in synthetic biology - the collective behaviour of synthetic microswimmers
Speaker: T. B. Liverpool, School of Mathematics, University of Bristol
April 4, 2018 at 10am in PAB 3-703
Abstract: I will present recent work studying the mechanisms of propulsion and the role of hydrodynamic interactions in the collective behaviour of collections of microscopic man-made catalytic active particles suspended in a fluid. I'll introduce a calculational framework that allows one to separate the different contributions to their collective dynamics from hydrodynamic interactions on different length scales. Hence one is able to quantify the effect of lubrication forces when the particles are very close to each other. I will show that they play as important a role as long-range hydrodynamic interactions in determining their many-body behaviour.
Title: Physical interactions and order in biological matter
Speaker: Kinjal Dasbiswas, University of Chicago
March 23, 2018 at 4pm in PAB 4-330
Abstract: There is accumulating recent evidence that physical interactions help to organize biological matter across scales from subcellular structures to cells and tissue. The statistical physics notion of order arising out of a competition of physical organizing forces and noise inherent in biology, provides insight into the physical mechanisms involved in as well as the dynamics of such ordering. Actively generated forces by molecular motors associated with the cellular cytoskeleton in particular can lead to long-range elastic interactions between cellular structures through mutual deformations of an underlying soft substrate. We first focus on how spatial order in heart muscle cells and their temporal beating are correlated through such mechanical factors. We then extend this picture to understand the recently observed ordered superstructures of molecular motors in non-muscle cells, underscoring the value of physical approaches in identifying general principles across different biological systems. Complementary to this elastic picture, other forms of order can arise in cytoskeletal components such as when they form a liquid crystal droplet phase discovered in recent experiments. The geometric effects inherent in such an anisotropic phase may also drive motor self-organization as well as droplet deformation. These ordered structures are biologically significant and play potentially important functional roles such as in muscle contraction and cell division. We conclude by noting that biology is inherently mechano-chemical and that mechanical interactions, usually faster and longer ranged than the diffusion of chemical signals, may also underlie pattern formation processes in biology.
Title: Statistical physics approach to study chromatin dynamics and nuclear organization
Speaker: Assaf Amitai, Massachusetts Institute of Technology
March 9, 2018 at 4pm in PAB 4-330
Abstract: Chromatin is a dynamical molecule which moves stochastically inside the cell nucleus. The encounter between chromatin sites is the rate-limiting step for many biological pathways. For example, double-strand DNA break repair by homologous recombination requires an efficient and timely search for a homologous template. This process can be observed with various microscopy techniques. I will discuss recent developments in polymer models, which combined with analyzing microscopy data, are used to infer the biophysical parameters of chromatin during the repair process.
The interaction of proteins with chromatin regulates many cellular functions. Most DNA-binding proteins interact both non-specifically and transiently with many chromatin sites, as well as specifically and more stably with cognate binding sites. These interactions and chromatin structure are important in governing protein dynamics. These questions can be addressed theoretically using diffusion models. I will show how that the dynamics of proteins is determined by the 3d organization of chromatin in the nucleus. The time to find a chromatin target depends on chromatin organization around it, which determines the local association and disassociation rates. Hence, the problem of facilitated search by a protein can be mapped to a continuous-time Markov chain. Protein motion in the nucleus is now being revealed through super-resolution microscopy. This approach would allow us to extract directly from microscopy data the interaction parameters of proteins at specific chromatin domains.
Title: Putting the mitotic spindle in its place
By Ehssan Nazockdast, Ph.D., University of North Carolina, Chapel Hill
March 7, 2018 at 1:00-2:00 PM in 5011 ENGINEERING V
Abstract: A cell is a complex fluidic environment in which many fundamental biological processes take place. One such is the proper positioning and elongation of the mitotic spindle which is crucial for chromosome segregation and asymmetric cell division, and involves the interaction of microtubule assemblies with motor-proteins and subcellular organelles. In a combined experimental and computational study, we use cytoplasmic flow measurements and computational fluid dynamics to argue that proper positioning is primarily achieved by the action of motor-proteins bound to the cell boundary.
Title: The Dynamics of Foraging and Starvation
By Sid Redner, Santa Fe Institute
February 9, 2018 at 4pm in PAB 4-330
Abstract: What is the fate of a random-walk forager that depletes its environment as it wanders? Whenever the forager lands on a food-containing site, all the food is consumed and the forager becomes fully sated. However, when the forager lands on an empty site, it moves one time unit closer to starvation. If the forager wanders S steps without encountering food, it starves to death. We show analytically that the lifetime of this starving random walk forager scales linearly with S in one dimension by solving an underlying non-Markovian first-passage problem. In greater than two dimensions, we present evidence that the lifetime grows quasi-exponentially in S.
We also investigate the role of greed, in which the forager preferentially moves towards food when faced with a choice of hopping to food or to an empty site in its local neighborhood. Paradoxically, the forager lifetime can have a non-monotonic dependence on greed, with different senses to the non-monotonicity in one and in two dimensions. In one dimension, the forager lifetime exhibits a huge peak when greed is negative, while in two dimensions the maximum lifetime occurs for positive, but not perfect, greed.
Title: “Mechanical models for collective behavior in epithelial tissues”
By Michael David Czajkowski, Syracuse University
January 26, 2018 at 4pm in PAB 4-330
Abstract: The mechanical properties of epithelial tissues, which are typically composed of a single layer of tightly bound cells, play an important role in development and disease. Therefore, our goal is to develop theoretical and computational tools to help explain global mechanical behavior and cell migration patterns in these tissues. In this talk, I will focus on connections between tissue rigidity and collective
migration using two different techniques in two different geometries. In the first project, we develop a hydrodynamic model for bulk tissues that couples cell shapes, tissue rigidity, and cell-substrate interactions by assuming that cells exert traction forces preferentially along gradients in the in-plane tissue stiffness, in accord with recent experiments. Analyzing the steady states of the model, we identify a new “morphotactic” parameter that our model predicts will control pattern formation in real tissues. In the second project, we developed a new active vertex model and employed it to simulate so called “wound-healing” experiments in which a tissue is allowed to expand into free space. These simulations indicate that tissues are unable to sustain a mechanical stress in the absence of propulsive forces generated by cellsubstrate interaction. Further, our results suggest that polar order in these propulsive forces qualitatively changes the observed stress profile, and may thereby promote cell proliferation.
Title: Constraints on cultures and characters
Professor Seppe Kuehn, University of Illinois at Urbana-Champaign
October 27, 2017 at 4pm in PAB 4-330
Abstract: Can we predict evolutionary and ecological dynamics in microbial communities? In this talk I argue that understanding constraints on biological systems provides a path forward to build predictive, phenomenological models. I present two vignettes which illustrate the power of elucidating constraints. First, in Nature microbial populations are subjected to variable levels of nutrients. Models of microbial growth stipulate that the rate of biomass production depends instantaneously on the levels of nutrients available. To test this assumption we subject populations of bacteria to periodic pulses of nutrients (feast) and long periods of starvation (famine) with varying frequency and amplitude. Exploiting precise measurements we show that that growth rate during feasts depends on the duration of famine. Our results are explained by a simple phenomenological model and have important implications for understanding abundance dynamics in microbial communities. In second study we ask how constraints on phenotypic variation limit the capacity of organisms to adapt to multiple simultaneous selection pressures. We select Escherichia coli for faster migration through a porous environment, a process which depends on both motility and growth. We find that a trade-off between swimming speed and growth rate constrains the evolution of faster migration. Evolving faster migration in rich medium results in slow growth and fast swimming, while evolution in minimal medium results in fast growth and slow swimming. A model of the evolutionary process shows that the genetic capacity of an organism to vary traits can qualitatively depend on its environment, which in turn alters its evolutionary trajectory [eLife, 2017]. We explore the possibility that phenotypic constraints and genetic architecture can provide a route to predicting evolutionary dynamics.
Title: Mathematical modeling of transport processes in cell biology
By Heather Zinn Brooks, University of Utah
October 13, 2017 at 4pm in PAB 4-330
Cells rely on combinations of diffusion and molecular motor transport of macromolecules for proper development and functionality. In this talk, I will present mathematical models of two scenarios where intracellular transport plays a key role. Using analysis of a model of partial differential equations, I show that synapse density maintenance in the ventral cord of C. elegans can be explained by a Turing-type mechanism, namely interactions of diffusion and actively transported proteins. I will then discuss a second scenario that shows the dark side of intracellular transport: molecular motors can be hijacked by viruses in order to travel more efficiently to the host cell's nucleus for replication. In this work, we derive an effective stochastic differential equation for viral transport to quantify first passage times to the nucleus.
Title: Antigen Affinity Discrimination in Germinal Center B-Cell Immune Synapses
Milos Knezevic, UCLA
October 6, 2017 at 4pm in PAB 4-330
Abstract: Efficient antibody responses need expansion of high affinity B cell clones. The initial stimulations of B cells occur when their B cell receptors (BCRs) bind to antigens found on the surfaces of antigen-presenting cells (APCs). These contacts lead to formation of immune synapses, which B cells use to extract antigen from the APCs. Typically B cells that express high-affinity BCRs obtain more antigen and compete better for T cell help. Immune synapses in naive and memory B cells form a large cluster of BCR-antigen complexes in the synapse center. Once this central cluster is assembled, B cells extract antigen from it via mechanical forces produced by the cortical cytoskeleton. However, in recent experiments it has been shown that germinal center (GC) B cells extract antigen by a distinct pathway which involves small peripheral clusters of BCR-antigen complexes. In this talk, we use Monte Carlo simulations to study pattern formation during GC B-cell – APC adhesion. We apply mechanical forces on BCR-antigen clusters above a certain threshold size, and investigate their effect on the quality of antigen affinity discrimination.
Title: Landau Theory for Assembly of Viral Capsids
Sanjay Dharmavaram, UCLA
September 22, 2017 at 4pm in PAB 4-330
The talk will discuss why the theory of the assembly of viral shells leads to fundamental problems with the Landau description of phase transitions. The talk will apply the Landau-Brazovskii theory for the solidification of chiral molecules on a spherical surface into an icosahedral density modulation. If assembly is based on assembling pentamer and hexamers then robust, weakly chiral icosahedral structures appear via first-order transitions. If it is based on the assembly of individual protein, then fragile icosahedral structures with a pronounced chiral character appear via weakly first-order or continuous transitions. Here, the icosahedral state is in close competition with states that have tetrahedral, octahedral and five-fold (D_5) symmetries. The application of the Landau theory of phase transitions to the problem of solidification on spherical surfaces requires allowing multiple irreducible representations of the symmetry group of the uniform phase to be realized at the transition point.
Encoding of distance, axis of travel, and topology at the juncture of hippocampal and posterior parietal systems for spatial mapping
Douglas Nitz, UC San Diego
June 20 2017, 4pm PAB 1-434
The subiculum and retrosplenial cortex together form a robust connection between hippocampal representations of position in the environment and posterior parietal cortex representations of position in route space. Our recent work examining the spatial firing properties of neurons within subiculum and retrosplenial cortex indicate that, in different ways, these structures encode conjunctions of spatial information concerning route spaces and their positioning and orientation in the larger environment. Furthermore, novel forms of spatial information encoding emerge in each structure. These forms include 'axis-tuning' and encoding of 'spatial analogies' in subiculum, and, in retrosplenial cortex, the encoding of position within route sub-spaces and distances between all route locations. The former are integrated temporally with the hippocampal CA1 sub-region through 'theta phase precession'. The latter, when considered in the context of retrosplenial encoding of head direction, could, in principle, form the basis for knowledge of overall path geometry. The results of these recent findings will be considered with respect to anatomical pathways that appear to represent the transformation of spatial cognition into action.
Formation of two-dimensional DNA films at the air/water interface
Jaime Ruiz-Garcia, Autonomous University of San Luis Potosí, Mexico
May 26, 2017, 4pm PAB 4-330
The central dogma in Langmuir monolayers studies is that they are formed when amphiphilic molecules are deposited and spread spontaneously on the water surface to form a monomolecular film. A typical characteristic of these systems is that they are formed by molecules with a hydrophilic head group and a large hydrophobic tail, such as fatty acids, phospholipids, etc. In this talk, we will show that monomolecular films at the air/water interface can also be formed using water soluble molecules. DNA is a highly charged polyelectrolyte and it is considered to be completely soluble in pure water. We show that DNA can be trapped at the air/water interface and does not go back into a water subphase. DNA is trapped in an energy minimum at the interface, much bigger that KBT, that does not permit its return to the water subphase. Once at the interface, DNA molecules condense to form two-dimensional foam-like mesostructures. This condensation occurs without the presence of multivalent cations. At high density, the molecules form a remarkable monomolecular network. At the interface, DNA is only partially immersed in water, which originates that the chains get only partially charged, but the charges are of the same sign. Therefore, this can be considered another case of like-charge attraction, similar to that found in colloids in restricted geometries such as the air/water interface and between glass plates. However, the origin of the attractive part of the interaction potential is unknown. In addition, we found that DNA at the air/water interface can form large 2D tetratic phase domains. Both the DNA monomolecular networks and the tetratic phase domains are interesting from theoretical and application standpoints.
Odd viscosity in chiral active fluids
Anton Sousolv, Leiden University
May 19, 2017, 4pm PAB 4-330
Abstract: Chiral active fluids are materials composed of self-spinning rotors that continuously inject energy and angular momentum at the microscale. Out-of-equilibrium fluids with active-rotor constituents have been experimentally realized using nanoscale biomolecular motors, microscale active colloids, or macroscale driven chiral grains. I will discuss how chiral active fluids break both parity and time-reversal symmetries in their steady states, giving rise to a dissipationless linear-response coefficient called odd (equivalently, Hall) viscosity in their constitutive relations. This odd viscosity provides no energy dissipation, but can give rise to a transverse flow – for example within a hypersonic shock. Such a viscosity term has been previously examined in the context of electron fluids subject to a magnetic field. However, only in active fluids does odd viscosity: (i) arise out of equilibrium, (ii) always come accompanied by an antisymmetric stress, and (iii) become ill-defined in the regime in which active rotations are hindered by interactions. I will examine the origins of odd viscosity and suggest how this property may be exploited to buildmachines powered by active fluids.
Optimizing self-assembly kinetics for biomolecules and complex nanostructures
William Jacobs, Harvard University
April 14, 2017, 4pm PAB 4-330
Abstract: In a heterogeneous self-assembling system, such as a large biomolecule or nanostructure, there is no guarantee that the lowest-free-energy state will form. Defects and mis-interactions among subunits often arise during a self-assembly reaction, particularly in systems comprising many distinct components. As a result, if we wish to assemble complex nanostructures reliably, we need to design robust kinetic pathways to the target structures, and not focus solely on their thermodynamic stabilities. In this talk, I shall describe a theoretical approach to predicting self-assembly pathways, with applications to both engineered nanostructures and natural biomolecules. First, I shall discuss design principles that can be used to tune the nucleation and growth rates of DNA 'bricks'. These principles have crucial implications for low-defect self-assembly and the design of time-dependent experimental protocols. Then, to highlight the biological importance of self-assembly kinetics, I shall present evidence that evolutionary selection has tuned ribosome translation rates to optimize the folding of globular proteins.
Dynamics of elastic waves on curved planar rods
Jonathan Matthew Kernes, UCLA
4pm PAB 4-330
Abstract: Lower dimensional elastic structures break isotropic symmetry of the material, introducing a preferred coordinate frame. Correspondingly, for an embedded structure the local tangential and normal components deformations respond in a separate but geometrically linked fashion. We exhibit the interaction between normal direction bending undulations, and tangential direction stretching deformations in the simplest case of a rod with local background radius of curvature. Curvature is found to decrease the availble oscillators in our system, as well as restrict the allowed frequencies.
Special Joint AMO/CM/CBP Seminar: Biomagnetic sensing with nitrogen vacancies in diamond
John Barry, Harvard University and MIT Lincoln Laboratory
March 3, 2017, 4pm PAB 1-434A
Abstract: Nitrogen vacancy (NV) color centers in diamond are rapidly emerging as a viable technology for quantum sensing. At the smallest scales, single NVs provide angstrom-scale spatial resolution with sensitivity sufficient for individual electron, proton, or protein detection. At much larger scales, bulk magnetometers harnessing large NV ensembles presently exhibit sensitivities surpassed only by SQUIDs and atomic vapor cells. Between these two limiting regimes, shallow surface layers of NVs allow for wide field-of-view magnetic imaging with diffraction-limited performance. The achieved combination of resolution and magnetic sensitivity remain unmatched for non-invasive imaging under ambient conditions, making solid state magnetic imaging favored for future investigations of various physical and biological phenomena. This talk presents progress towards one primary application: magnetic detection and imaging of action potentials from single neurons. We demonstrate this method using excised axons from two invertebrate species, marine worm and squid; and then by single-neuron action potential magnetic sensing exterior to intact, live, opaque marine worms. Extended-duration magnetic sensing is performed with no adverse effect. We discuss future steps to enable non-invasive imaging of functional mammalian neural networks in real time.
Statistical physics of molecular evolution: from gene regulation to the immune system
Armita Nourmohammad, Princeton University
March 3, 2017, 3pm PAB 4-330
A venerable question in evolutionary biology is: if the tape of life was replayed, would the outcome be the same? We do not know how evolutionary predictability relates to different molecular scales, ranging from genotypes (DNA and amino acids) to molecular phenotypes (functions such as protein activity). I discuss universal properties of molecular phenotypes, encoded by genotypes with large degrees of freedom, which allow for the predictive description of their evolution. Populations are often subject to time-dependent pressure from the environment. I introduce a non-equilibrium framework for adaptive evolution of molecular phenotypes in time-dependent conditions. As an example, I present strong evidence that environmental fluctuations drive the evolution of gene expression levels in Drosophila. Co-evolving populations reciprocally affect the fitness of each other, acting as time-dependent environments with feedback. I show evidence of co-adaptation between interacting cellular populations of HIV viruses and the antibody repertoire of a patient over the course of an infection. In particular, I discuss the conditions for emergence of broadly neutralizing antibodies, which are recognized as critical for designing an effective vaccine against HIV.
Step-by-step shape evolution of tubular solids by defect motion
Daniel Beller, Harvard University
February 24, 2017, 4:00pm PAB 4-330
Abstract: Two-dimensional crystalline order on surfaces with cylindrical topology gives rise to helical lattices. This type of packing occurs in biology at many scales, from biofilaments to viral capsids to botany, as well as in carbon nanotubes and in colloidal crystals. Shape changes in the tubular surface generally require changes in the crystalline tessellation of the surface. This evolution can be undertaken step-by-step via the motion of elementary defect pairs through the tubular crystal. I will discuss the physics of plastic deformation in tubular crystals by the unbinding and glide separation of dislocation pairs. Through theory and simulation, this work examines how the tube’s radius and helicity affect, and are in turn altered by, the mechanics of dislocation glide. The system’s bending rigidity plays an important role, and can resist, arrest, or even reverse the deformations of tubes with small radii. I will also discuss the equilibrium shapes of tubes containing dislocations, with potential implications for biofilaments such as microtubules.
Activity, specificity, selectivity: new approaches for micro-assembly
Carl Goodrich, Harvard University
February 10, 2017, 4:00pm PAB 4-330
Abstract: Biology is able to build remarkable machines through a combination of programmable interactions and specific energy input through small molecules (ATP). However, we are largely unable to manufacture artificial structures of comparable complexity at the microscopic scale. The development of specific DNA glues has revolutionized our ability to program interactions, but we have barely scratched the surface of what can be built. I will discuss new approaches for micro-assembly, focusing on the use of self-propelled particles (a potential alternative to ATP) to drive out-of-equilibrium assembly. Under the right conditions, for example, the motion of active colloids can be programmed to pull semiflexible filaments into topological structures such as braids and weaves. I will also discuss how DNA interactions can be used to program a selectively permeable gel.
Aggregation of proteins: growth of glucagon fibrils and bacterial growth
Andrej Kosmrlj, Princeton University
February 3, 2017, 4:00pm PAB 4-330
Abstract: Misfolding and aggregation of peptides and proteins are the hallmarks of many human diseases. With the advancement of microscopes, it is now possible to observe the kinetics of individual aggregates and fibrils in vitro. Interestingly, in some cases the growth of fibrils is intermittent, where the periods of growth are interrupted by periods of stasis. In this talk I will focus on the intermittent fibrillation of glucagon and I will describe how E. coli bacteria deal with harmful aggregates of misfolded proteins. Glucagon is a peptide hormone that upregulates blood sugar levels and is used to treat diabetic patients in situations of acute hypoglycemia. When dissolved in a fluid state, glucagon can form fibrils and become useless, as the fibrils cannot be absorbed and used by the body. The observed intermittent growth of glucagon fibrils can be explained with a simple model, where fibrils come in two forms, one built entirely from glucagon monomers and one entirely from glucagon trimers. The opposite building blocks act as fibril growth blockers, and this generic model reproduces experimental behavior well. Finally, I will discuss how E. coli bacteria deal with harmful aggregates of misfolded proteins that develop, when bacteria are under heat or antibiotic stress. In order to maximize the fitness of the whole population, bacteria distribute aggregates asymmetrically between their daughters, such that one daughter inherits the whole aggregate, while the other daughter receives none of it. Over time such asymmetric distribution of aggregates produces many “rejuvenated” bacteria with small aggregates that are quickly dividing at the expense of a few bacteria with large aggregates that are dividing very slowly.
Speaker: Tim Stutz and Alfonso Landeros (UCLA)
Date: Friday, January 25, 2019 at 4:00PM in PAB 4-330