Student Poster Abstracts

Programmable icosahedral capsids: A layered approach mediated by lipid templates

Self-assembly is a fundamental process in both inanimate and animate systems, where subunits spontaneously organize into structured assemblies. Spanning from lipid membranes to multiprotein filaments, this self-organization is crucial for numerous biological functions. A striking example is the assembly of protein subunits around viral nucleic acids to form protective capsids. Over 50 years ago, the Caspar and Klug theory of quasi-equivalence provided seminal insight into viral capsid structure, observing that most spherical viruses exhibit icosahedral order. While this theory describes the structure of many smaller spherical viruses, larger viruses such as herpes simplex virus assemble in layers and require a template for successful assembly.

Despite extensive studies on the spontaneous assembly of viral proteins into empty capsid shells, the in vitro replication of templated assembly around a rigid spherical template remains unexplored. Here, we aim to construct large capsids with high yield using a minimal number of building blocks, inspired by the layered architecture of large viruses. By utilizing a spherical template to assemble capsids with a single equivalent bonding interaction, we propose an experimental strategy that mimics this natural design. This approach employs engineered lipid vesicles with cholesterol-modified DNA as the inner layer and triangular-shaped DNA origami building blocks as the outer layer.

We developed a single-molecule microscopy assay to investigate the kinetics of DNA origami self-assembly on lipid bilayers using Total Internal Reflection Fluorescence (TIRF) microscopy. Furthermore, we devised a method to obtain quantitative estimates of triangle-lipid binding interactions through flow cytometry on lipid-coated colloids. This comprehensive approach promises to significantly enhance the yield of origami capsids, potentially by several orders of magnitude compared to current methods.

Universal scaling of shear thickening suspensions under acoustic perturbation

Nearly all dense suspensions undergo dramatic and abrupt thickening transitions in their flow behavior when sheared at high stresses. Such transitions occur when suspended particles come into frictional contact with each other to form structures that resist the flow. These frictional contacts can be disrupted with acoustic perturbations, thereby lowering the suspension’s viscosity. Acoustic perturbations offer a convenient way to control the suspension’s shear thickening behavior in real time, as the suspension responds to the perturbation nearly instantaneously. Here, we fold these acoustic perturbations into a universal scaling framework for shear thickening, in which the viscosity is described by a crossover scaling function from the frictionless jamming point to a frictional shear jamming critical point. We test this theory on sheared suspensions with acoustic perturbations and find experimentally that the data for all shear stresses, volume fractions, and acoustic powers can be collapsed onto a single universal curve. Within this framework, a scaling parameter that is a function of stress, volume fraction and acoustic power determines the proximity of the system to the frictional shear jamming critical point and ultimately the viscosity. Our results demonstrate the broad applicability of the scaling framework, its utility for experimentally manipulating the system, and open the door to importing the vast theoretical machinery developed to understand equilibrium critical phenomena to elucidate fundamental physical aspects of the non-equilibrium shear thickening transition.

Phase Separation driven by Active Flows

We study the hydrodynamics of an active nematic fluid mixed with another passive or active fluid. We model this by extending the continuum theories of active nemato-hydrodynamics to a two-fluid mixture with separate velocity fields for each fluid component, coupled through a viscous drag. We observe phase separation, and argue that this results from an interplay between active anchoring and active flows driven by concentration gradients. The results are an example of liquid-liquid phase separation out of equilibrium, and may be relevant to cell sorting.

Formation of motile cell clusters in heterogeneous model tumors: the role of cell-cell alignment

The formation of migrating cell clusters plays an important role in many biophysical processes, including cancer metastasis. In this work, we investigate how a potential cell-cell alignment mechanism affects the formation of motile cell clusters in the bulk of a heterogeneous tumor. We model the tumor as a two-dimensional heterogeneous confluent layer with both motile and non-motile cells using the Cellular Potts formalism. Our results indicate that the degree of clustering is governed by two distinct processes: the formation of clusters due to the presence of cell-cell alignment interactions among motile cells, and the suppression of clustering due to the presence of the dynamic cellular environment (comprised of the non-motile cells).

Kinetic trapping determines actin filament organization, thereby controlling the shape of liquid droplets

Biomolecular condensates are known to act as concentrators of essential proteins resulting in amplification of cellular signaling. The role of condensates in controlling the dynamics of cytoskeleton is not fully understood. Recent evidence suggests Vasodilator stimulated phosphoprotein (VASP) forms liquid droplets and can nucleate, polymerize, and bundle actin filaments to enclose shell and ring-shaped actin networks (Graham et al., Nature Physics, 19, 574–585 (2023)). To understand the mechanism of formation, we used CytoSim, an agent-based actin network simulator and studied the role of VASP-actin interactions under hard wall boundary conditions (https://doi.org/10.1101/2023.05.26.542517). Our simulations reveal that (un)binding kinetics determine the probability of shell vs. ring formation consistent with kinetically trapped structures. We also show that reduction of Actin-VASP residence time would favor shell formation and also establish it with experimental studies. Additionally, we show that rings are only formed under a small subset of conditions studied that provide effective bundling and form at a higher probability under slower filament elongation rates. To highlight the role of nucleation, we study the competition between Arp2/3 and VASP to show that Arp2/3 inhibits ring formation through nucleation of actin filaments (https://doi.org/10.1101/2023.06.23.546267). Experimental evidence also suggests that spherical liquid droplets deform to form ellipsoidal droplets explained by the balance between actin bending and droplet surface energies. We study the reorganization within our simulated droplets containing actin rings and show that filament length plays a critical role in determining whether the ring would eventually become a linear bundle (sphere-to-rod) or if they would remain kinetically trapped in a ring-like structure (sphere-to-ellipsoid). Our findings are relevant to the understanding of the kinetic and mechanical principles that guide the organization of filamentous actin networks within phase-separated liquid droplets.

Title to be announced

Abstract to come.

Mechanical Insights into Multicomponent Coexistence: A General Theory for Nonequilibrium Systems

Nonequilibrium multicomponent systems, such as non-reciprocally interacting systems and mixtures of active and passive particles, routinely experience phase coexistence. In equilibrium, thermodynamics provides a framework to determine the phase diagram of these systems solely from measurable, bulk equations of state. Out of equilibrium, however, it is unclear if phase diagrams can always be constructed solely from bulk equations of state. Here, we leverage balance laws to develop a theory of multicomponent coexistence that is purely mechanical, finding a framework akin to thermodynamics under certain conditions. We find general bulk criteria for coexistence and demonstrate when an effective equilibrium form is admitted. Lastly, we numerically validate our theory on common models of nonequilibrium multicomponent field dynamics.

Mechanical Heterogeneity in Fish Embryos

Tissue mechanics is essential for animal development, yet the precise manner in which mechanics determine the fate of embryonic tissue is not fully understood. Our system of interest is the enveloping cell layer (EVL) of an Annual Killifish embryo during the epiboly stage (11-22 hpf), at which the EVL exhibits a wide and heterogeneous apical cell area size distribution. In this study, we employ a technique called micromechanics to infer mechanical fields, explore tissue heterogeneity, and construct quantitative-based models directly from experimental images. Applying micromechanics to a vertex model, we demonstrate that the experimental area distribution can only be reproduced with a tissue exhibiting heterogeneous mechanical properties. The heterogeneity in the model tissue is characterized by larger cells being softer than smaller ones. This correlation allows us to write functional relations between the model parameters and the experimental tissue geometry, which were used to construct new energy functional candidates for the tissue. Simulations suggest that promoting cell intrusion via chemical activation is a viable experimental assay to confirm the heterogeneous properties of embryos. Here, the area strain versus cellular area is the critical observable quantity, where, counterintuitively, a heterogeneous tissue should exhibit a flat response while in a homogeneous tissue, larger cells should deform more. We discuss the potential implications of these heterogeneities on the fate of the tissue and how they could act to reduce stress on the cell layer, potentially serving as an important evolutionary mechanism for the survival of the animal. Additionally, micromechanics can be applied to a wide range of experiments and models, making it a valuable technique for future studies

Optimal Cell Migration in Heterogenous Environments

Collective cell migration is an important phenomenon in many biological processes, such as embryonic development, cancer metastasis, and tissue repair. One of the key features of this is the directionality and coordination of the multicellular movements through some biomechanical interactions that are not well understood yet. To understand the role of tissue dynamics (rigidity, interfacial tensions, and differential activity) on the migration of cellular clusters, we employed a 2D Self-propelled Voronoi model (cell motility is modelled via the active ornstein-uhlenbeck process) coupled with an additional interfacial tension controlling the heterotypic cell interactions to simulate a cell cluster migration in a confluent tissue. We find that a caged-to-migratory transition happens in monolayers (homogeneous in rigidity) for increasing cluster activity very similar to the activity-induced solid-fluid transition that is a universal feature of these confluent models. Interfacial tension, on the other hand, regulates the cluster integrity during the movement and invokes leader-follower like dynamics in mixed clusters of varying cell motilities. Both in the cases of pure motile cell clusters and mixed motile and non-motile cell clusters, we observe optimality in the Heterotypic Interfacial Tension (HIT) strengths for varying cluster sizes and motilities i.e low HIT strength isn’t enough for coherent migration and high HIT strength leads to a pinning-like effect at the interface between different cell types caging the cells from movement. To test the effectivity of this 2D Voronoi modelling framework, we focused on simulating cell clusters with several components (different activity levels and interfacial tensions with each other) to mimic the coherent ingression of cells with varying Nodal levels during zebrafish gastrulation. By introducing Nodal-dependant values for cell motility and differential heterotypic interfacial tensions (DHIT), we are able to correlate the experimental observation qualitatively to our model predictions, highlighting the importance of the interplay between cell adhesion and directed motility for efficient cluster migration.

Self-learning mechanical circuits

Dynamical self-adaptivity and embodied computation are ubiquitous in living systems, from cytoplasm and biofilms to animal flocks. Despite recent progress in such material computation, the problem of designing mechanical systems which self-learn remains poorly understood. Here we introduce the concept of self-learning mechanical circuits, which use mechanical inputs from changing environments to continually update their internal state, thus representing an entirely mechanical information processing unit. Our circuits are composed of a new mechanical construct: an adaptive directed spring (ADS), which executes neural network-like computations mechanically. We demonstrate the capacity of mechanical circuits to self-learn using both theoretical modeling and experimental realizations. Our results pave the way towards the construction of energy-harvesting, adaptive materials which can autonomously and continuously sense and self-optimize to gain function in different environments.

The Non-Equilibrium Statistical Embryo

How do collectively organized multi-cellular life forms emerge from the exchange of individual mechanical and morphogenetic cues? Deciphering multiscale population dynamics poses a fundamental challenge to understanding the mechanism whereby embryonic growth is coordinated on larger scales. Coupling high-resolution optical imaging with non-equilibrium statistical analysis provides phenomenological insight into the role of mechanics and shape on macroscopic ordering during embryogenesis. Establishment and maintenance of asymmetric pre-patterning in the C. elegans embryo is an ideal subject for statistical exploration of self-assembly, however a physical theory of development from germ layer specification through morphogenesis is lacking. Analyzing stochastic trajectories according to cell fate distinctions and generation dependent control was found to be a robust statistical strategy for defining transition points at which collective space-time morphological patterns emerge. Significantly, oscillations in anomalous diffusion during gastrulation suggest that periods of relaxation accompany waves of collective motion and an increase in super diffusive behavior from differentiated cells suggests that non-equilibrium activity shifts from being driven by proliferation to directed motion. Our data-driven theoretical approach identifies self-organization principles governing proliferation in confinement and yields a new perspective about the embryo as active matter.

Hyperuniformity at the Absorbing State Transition

Hyperuniform systems, whereby the static structure factor (or density correlator) obeys S(q)~q^h with h > 0, emerges at criticality in systems having multiple absorbing states, such as periodically sheared suspensions. These lie in the conserved directed percolation (C-DP) universality class, for which analytic results for h are lacking. Here, using Doi-Peliti field theory for interacting particle systems and performing perturbative renormalization group about a Gaussian model, we find h=0+ and h=2ε/9+O(ε^2) in dimension d > 4 and d = 4−ε respectively. We show how hyperuniformity emerges from anticorrelation of strongly fluctuating active and passive densities, and highlight the significance of an ‘irrelevant’ conserved noise.

The Assembly and Budding of SARS-CoV-2

Upon the accumulation of viral structural proteins along the ER-Golgi intermediate compartment (ERGIC), SARS-CoV-2 assembles and buds off, driven by the interactions between these proteins, RNA and the ERGIC membrane. The membrane or M protein is thought to recruit other structural proteins, leading to the formation of protein aggregates and the subsequent induction of membrane curvature, prompting the onset of virion formation. However, the direct impact M protein has on the membrane and how this leads to assembly is unclear. Here, we combine all atom molecular dynamics (MD) simulations of an individual M protein with a mesoscopic continuum model describing the coupled evolution of membrane shape and M protein density to quantify viral assembly and budding. From our MD simulations, we identify the M protein’s ability to thin the membrane and induce curvature dependent on its conformation. By incorporating these properties into our continuum model and then comparing with atomic force microscopy measurements of protein aggregate formation, we estimate the membrane mediated M-M protein interaction and make predictions for the onset of assembly and budding under physiological conditions. This work provides a better understanding of how the interactions and dynamics of M protein lead to viral assembly and budding, supplying insights into alternative methods for preventing viral replication and implications for other enveloped viruses.

Exceptions to the ratchet theorem and their marginal stability

Breaking time-reversal and left-right symmetries in stochastic dynamics generically leads to steady-state density currents (ratchet currents) – a principle known as the “Ratchet Theorem.” In active matter, a typical example is active particles in an asymmetric potential. An interesting exception is non-interacting self-propelled particles in an asymmetric activity landscape, which does not lead to steady currents. Surprisingly, the ratchet current is restored by including interactions between the particles. We show this exception to the ratchet theorem (among others) possesses a hidden time-reversal symmetry, which is marginally stable. We explain the interaction-induced ratchet current through a mean-field picture.

Liquid Crystal Order of Cells near Corners

Anisotropically shaped cells tend to self-align, giving rise to domains of nematic ordering while experiencing splay and bend, and forming topological defects akin to those observed in 2D nematic liquid crystals. To elucidate the physical properties of this unique liquid crystal, the research explores the behavior of monolayers of cells in proximity to corners and sharp edges. Through in-vitro experiments, a distinct correlation between wedge angle and the nature of deformation in cell monolayers becomes apparent. Smaller wedge angles predominantly trigger splay deformation, whereas larger angles induce bend deformation. Notably, the angle at which splay and bend deformations are equally likely is determined by the ratio between splay and bend elastic constants and becomes thus an indirect measurement of the elastic anisotropy of the system. Furthermore, the splay and bend deformations under confinement are influenced by the adhesion strength of the cells with the substrate. This investigation offers valuable insights into the intricate interplay among cellular morphology, confining geometry, and adhesion properties, shedding light on the mechanics of cellular self-alignment and deformation.

Controlling wall particle interactions with

Abstract to come.

Modelling A Growing Active Nematic That Collectively Senses an Anisotropic Substrate

Apolar elongated cells growing on an aligned liquid crystal elastomer collectively sense the substrate orientation. As cells grow and divide along their long axis, they create locally aligned domains of high density separated by low density disordered regions. The ordered domains then grow in size and align with the substrate. At high densities cells jam and growth ceases. The jamming density depends on the order: Highly aligned cells jam at higher densities. The degree of order of the final state depends on the initial seeding density, with smaller seeding densities leading to more aligned monolayers. While the mechanism of sensing the substrate orientation is not known, adding a focal adhesion kinase inhibitor, which disrupts cell-substrate adhesion, leads to the loss of collective directional sensing. We model this system as an overdamped compressible active nematic with growing density. The initial cell aggregation is driven by a Cahn-Hilliard like free-energy and the coupling of density gradients to the orientational dynamics. We incorporate the interaction with the substrate via two possible mechanisms: as an external field biasing the direction of the nematic, and as anisotropic friction experienced by cells. We first show that arresting rearrangements and growth by making rotational viscosity and growth-rate density dependent reproduces the dependence of final order on seeding density. We next show that activity, anisotropic friction, flow-alignment, and alignment of Q-tensor to density gradients all act in concert to engender substrate-sensing at the collective level. We qualitatively reproduce the swirling and laning patterns seen in experiments and plan on exploring the relative importance of friction anisotropy, activity, and density-gradient alignment.

Using optogenetics to disentangle how mechanical cues coordinate cell behaviors and maintain bilateral symmetry during tissue elongation

During Drosophila axis elongation, both global embryo-scale forces and local cell-scale forces contribute to the elongation of the germband epithelium along the anterior-posterior axis of the embryo. Divided by the invaginated mesoderm, the left and right sides of the germband extend to the same length and display a remarkable degree of symmetry at the cellular scale in terms of rearrangement rates. The co-elongation of neighboring tissues, which is a conserved symmetry-breaking mechanism in many embryos, is an understudied but interesting phenomenon that requires coordination of cell behaviors across embryonic length scales. With optogenetics, we can disentangle these relationships by inducing spatiotemporally defined biomechanical perturbations in vivo in a way that is not feasible via pharmacological or genetic interventions. Using previously described optogenetic tools that target the Rho/Rho-kinase signaling pathway and actomyosin force generation, we demonstrate a protocol for corralling optogenetic perturbation to either the left or right side of the germband epithelium while simultaneously monitoring the effects on actomyosin and cell behaviors on both sides of the embryo. Notably, we find that local optogenetic perturbation leads to disruption of myosin recruitment and planar polarity within the activated side of the embryo as well as unexpected effects in distant, unactivated regions of the embryo, such as prominent dorsal folds, which suggests coupling that contributes to overall axis elongation. These studies provide a foundation for unraveling how mechanical interaction and feedback help coordinate cell behaviors, even over long distances, to ensure robust morphogenesis and development.

Beyond 2D Tissue Mechanics: A 3D Continuum Shell Theory for Epithelial Cells

Two-dimensional (2D) models for confluent tissues correlate the mechanical state of monolayers to morphological indicators, such as the shape of their cellular components. However, this approach proves inadequate for in-vitro cultured MDCK epithelia. Confocal microscopy reveals significant and systematic variations of cross-sectional cell shape along the apical-basal axis, contradicting 2D theory even when averaged. Therefore, we model the shape of the cell boundary in three dimensions using elastic shell theory with appropriate boundary conditions directly related to the cellular actin distribution, particularly the fiber bundles at the basal side. The analytical solutions align with our experimental observations and allow realistic modeling of actin-related cell mechanics based on observable biological effects. This 3D continuum shell model serves as a rich platform for deriving effective morphological indicators aimed at diagnosing tissue mechanics and health.

Hydrodynamic force on lipid-anchored proteins depends on protein shape

Flow in the surrounding fluid can transport lipid-anchored proteins laterally across lipid membranes. This transport can separate proteins according to their folded size and shape, since larger proteins experience higher force from shear flow. Here we present microfluidics-based measurements of protein transport across supported lipid bilayers. Our experiments use fluorescence microscopy to detect femtonewton-scale forces on the aqueous part of the protein with sufficient precision to distinguish between proteins with similar molecular weights but different shapes.

Emerging States in Dense Systems of Chiral Active Particles

Across scales, from molecules to tissues, dense biological systems can exhibit collective dynamics driven by activity and elastic interactions, including flocking transitions and long-range spatiotemporal order. In this work, we consider densely packed systems of active Brownian particles with intrinsic individual chirality from a theoretical perspective, showing that these can lead to the emergence of a variety of states, including collective rotating mesoscopic order. Using dual analytical approaches—one based on normal modes and the other on continuum elasticity—this study provides a comprehensive understanding of the states that can be observed in the systems, matching very well our numerical simulations. Our findings suggest that the collective rotating states that we identify may generically appear in natural and artificial dense chiral active systems.

Crumpled thin sheets – a tabletop journey through nonequilibrium glassy dynamics

Understanding the unusual but seemingly universal dynamics of disordered systems trapped far from equilibrium remains a major challenge in condensed matter physics. Yet, they are not very hard to observe. Take, for example, a thin plastic sheet and crumple it into a ball. It might not be immediately evident, but this seemingly mundane object exhibits many of the hallmark behaviors shared by complex non-equilibrium and disordered systems. These include slow relaxations and creep, intermittent mechanical responses, emission of broadly distributed crackling noise and a range of memory effects.

The macroscopic tabletop nature of crumpled sheets allows us to measure and correlate observables at all scales – from changes in the local structure to global mechanical response. We find that the effective degrees are freedom are localized, bistable elements, similarly to many amorphous solids. We then leverage unique experimental accessibility to build to bottom-up understanding of complex macroscopic phenomena. Under cyclic strain, the bistable elements to emergent memory effects and a programmable stiffness. Under constant loading, their slow yet correlated activation gives rise to logarithmic creep and a beautiful example of Self-Organized Criticality. We identify the minimal underlying principles for these phenomena such that our insights can be exported to disordered systems on all scales, from supercooled liquids to earthquake dynamics.

Self-organisation of mortal filaments and its role in bacterial division ring formation

Filaments in the cell commonly treadmill. Driven by energy consumption, they grow on one end while shrinking on the other, causing filaments to appear motile even though individual proteins remain static. This process is characteristic of cytoskeletal filaments and leads to collective filament self-organization. Here we show that treadmilling drives filament nematic ordering by dissolving misaligned filaments. Taking the bacterial FtsZ protein involved in cell division as an example, we show that this mechanism aligns FtsZ filaments in vitro and drives the formation of the division ring in living Bacillus subtilis cells. We find that ordering via local dissolution also allows the system to quickly respond to chemical and geometrical biases in the cell, enabling us to quantitatively explain the ring formation dynamics in vivo. Beyond FtsZ and other cytoskeletal filaments, our study identifies a mechanism for self-organisation via constant birth and death of energy-consuming filaments.

Critical Phenomena in Biomolecular Condensates: Structure and Dynamics

The study of biomolecular condensates in a cell has become increasingly important due to its implications in health physiology and disease. Phase separation has become a popular framework to study them. Any system which shows phase separation can be characterized by an underlying phase diagram, with tie lines. What implications does location on the phase diagram have on the properties of the biomolecular condensates? To address this question, we focus on a fascinating point on the phase diagram – the critical point and seek to enumerate all the interesting effects related to proximity to the critical point. We start with a very simple model system for biomolecular condensates – mixture of protein (BSA) in polymer (PEG) whose phase diagram is well known to us. We measure the supramolecular structure and material properties of the protein mixtures near the critical point. We compare to scaling theories and discuss implications for the  dynamics of condensates.

Tunable Wetting Properties in Multicomponent Protein Condensates

Drawing inspiration from experimental studies at Wei lab, we explore the tunable wetting properties of multi-component protein condensates using computer simulations and a sticker-spacer model design. Our key discovery is that the wetting behavior—specifically how protein droplets interact—can be adjusted by the strength of weak, non-specific interactions within a client protein present in the droplets. Weak interactions promote droplet contact, whereas their absence results in non-wetting behavior. Strong interactions, on the other hand, lead to droplet coalescence. Additionally, the volume fractions of these components significantly influence the resulting structures.

Sterol-lipids enable large-scale, liquid-liquid phase separation in membranes of only two components

A wide diversity of membranes, from those with hundreds of lipids (as in vacuoles of living yeast cells) to as few as three (as in artificial vesicles) phase separate into micron-scale liquid domains. This limit of three components is perplexing from a theoretical standpoint: only two components should be necessary. It is equally perplexing from an experimental standpoint: only two lipid types are required to form large-scale liquid domains in lipid monolayers. This incongruity inspired us to search for single, joined “sterol-lipid” molecules that replace both a sterol and a phospholipid in membranes undergoing liquid-liquid phase separation. By using sterol-lipids with long, saturated chains, we sought to mimic known preferential interactions between cholesterol and lipids with high melting temperatures. We find that membranes with only two components (one of which is a sterol-lipid) do indeed phase-separate into micron-scale liquid domains. This result mitigates experimental challenges in determining tie-lines and in maintaining constant chemical potentials of lipids as lipid ratios are changed. For one of the binary membranes, we construct a miscibility phase diagram to show how the membrane’s phase separation depends on temperature and the ratio of the lipids.

Linker-mediated designed self-assembly and superselectivity

Using multiple weak bindings to form effectively strong interactions, namely multivalent interactions, constitutes the cornerstone of various biological processes. Compared to conventional multivalent bindings, indirect linker-mediated bridging provides new axes for regulating interactions by adjusting both the density and specificity of linkers. In this context, linkers are defined as small, dual-ended molecules capable of transiently and reversibly binding to receptors and ligands on host and guest structures, respectively. This creates bridges that connect these entities. The specificity of the linkers also enables in situ programmability, allowing for the precise “stitching” together of building blocks in synthetic multicomponent systems by introducing specific linkers.

I will present our recent work on the linker-mediated designed assembly of DNA-coated colloids and the superselectivity observed in linker-mediated multivalent nanoparticle adsorption. I will discuss the intriguing entropy-induced phase behaviors resulting from both valence-limited and multivalent natures. Additionally, I will highlight the potential application of the designed precise fabrication on nanoscale, as well as the design of new sensors to detect biomarkers.

Finding signatures of low-dimensional geometric landscapes in high-dimensional cell fate transitions

In many animals, hundreds of highly specialized cell types work together to maintain homeostasis. When growing or injured, cells can self-organize and transition between these cell types. The consistency and robustness of developmental cell fate trajectories suggests that complex gene regulatory networks effectively act as low-dimensional cell fate landscapes. While there has been progress in characterizing classes of cell fate decisions using gradient-like dynamical systems, the theory connecting geometric landscapes to high-dimensional gene expression space is still in its infancy. In this paper, we introduce a phenomenological model of cell fate transitions that predicts bifurcation signatures observable in gene expression measurements. By combining low-dimensional gradient dynamical systems and high-dimensional Hopfield networks, our model captures the interplay between cell fate, gene expression, and signals. The signal-driven bifurcations of an input landscape control the stability of attractors, and this feature allows the model to predict dynamics resulting from any class of bifurcation. Using existing single-cell RNA-sequencing time-series data, we compare experimental observations to theoretical landscape candidates belonging to different bifurcation classes. In the developing mouse lung, the transient appearance of a mixed alveolar type 1/type 2 state in the growing mouse suggests the maturation of alveolar cells is a triple cusp bifurcation. Additionally, previous analysis of lineage-tracing data of in vitro hematopoetic differentiation indicated that monocytes have a neutrophil-like path of differentiation; when compared to possible landscapes, bipotent neutrophil-monocyte progenitors appear to undergo a heteroclinic flip bifurcation. These results show that a geometric landscape approach can reveal new insights in time series single-cell RNA-sequencing data of cell fate transitions.