51�������

Abstract:

Differential equations containing memory terms that depend nonlinearly on past states model a variety of non-Markovian processes. In this study, we present a Markovian embedding procedure for a subclass of such equations with distributed delay by utilising an exact spectral representation of the nonlinear memory function. This allows us to transform the nonlocal system to an equivalent local-in-time system in an abstract extended space. We demonstrate our embedding procedure for two one-dimensional physical models: (i) the walking droplet and (ii) the single-phase Stefan problem. In addition to providing an alternative representation of the underlying physical system, the local representation finds applications in designing efficient time-integrators with time-independent computational costs for memory-dependent systems which typically suffer from growing-in-time costs.

Abstract:

Most living things on Earth - from bacteria to humans - must migrate in some way to find favourable conditions. Therefore, they nearly all use chemotaxis, in which their movement is steered by a gradient of chemicals. Chemotaxis is fundamental to many processes that control our well-being, including inflammation, neuronal patterning, wound healing, tumour spread in cancer, even embryogenesis. Understanding it is a key goal for biologists. Despite the fact that many basic principles appear to have been conserved throughout evolution, most research has focused on understanding the molecular mechanisms that control signal processing and locomotion. Cell signaling - cells responding to time-varying external signals - underlies almost all biological processes at the cellular scale. Chemotaxis of single cells provides particularly amenable model systems for quantitative cell signaling studies, even in the presence of noise and fluctuations, because the output, the cell's motility response, is directly observable. However, the different scientific disciplines involved in chemotaxis research rarely overlap, so biologists, physicists and mathematicians interact far too infrequently, methodologies and models differ and commonalities are often overlooked, such as the possible influence of physical or environmental conditions, which has been largely neglected.

Abstract:

We devise tractable models of unitary quantum many-body dynamics on tree graphs, as a first step toward a deeper understanding of dynamics in non-Euclidean spaces. To this end, we first demonstrate how to construct strictly local quantum circuits that preserve the symmetries of trees, such that their dynamical light cones grow isotropically. For trees with coordination number 𝑧, such circuits can be built from 𝑧-site gates. We then introduce a family of gates for which the dynamics are exactly solvable; these satisfy a set of constraints that we term “tree-unitarity.” Notably, tree-unitarity reduces to the previously established notion of dual-unitarity for 𝑧 =2, when the tree reduces to a line. Among the unexpected features of tree-unitarity is a trade-off between “maximum butterfly velocity” dynamics of out-of-time-order correlators and the existence of nonvanishing correlation functions in multiple directions, a tension absent in one-dimensional dual-unitary models and their Euclidean generalizations. We connect the existence of (a wide class of) solvable dynamics with nonmaximal butterfly velocity directly to a property of the underlying circuit geometry called 𝛿-hyperbolicity, and argue that such dynamics can only arise in non-Euclidean geometries. We give various examples of tree-unitary gates, discuss dynamical correlations, out-of-time-order correlators, and entanglement growth, and show that the kicked Ising model on a tree is a physically motivated example of maximum-velocity tree-unitary dynamics.

Abstract:

We develop a continuum theory to describe the collective dynamics of deformable epithelial cells, distinguishing the force-generating active filaments in the cells from their shape. The theory demonstrates how active flows driven by active filaments can create nematic domains and topological defects in the cell shape field. We highlight the role of the filament flow-aligning parameter, λQ, a rheological quantity that determines the response of the filaments to velocity gradients in the active flows, and plays a significant, to date unappreciated, role in determining the pattern of extensional and compressional active flows. In a contractile cell layer, local flows are expected to align elongated cells perpendicular to the active filaments. However, with increasing λQ, long-range correlations in the active turbulent flow field lead to extended regions where this alignment is parallel, consistent with recent experiments on confluent Madin-Darby canine kidney (MDCK) cell layers. Further, we distinguish defects in the filament director field, which contribute to the active driving, and those in the shape director field, measured in experiments, which are advected by the active flows. By considering the shape-filament orientation, we explain the unexpected motion of +1/2 defects towards their head in contractile cell layers, consistent with recent experiments on epithelial layers examining stress around shape defects.