Publications

Few techniques are available for studying the nature of forces that drive subcellular dynamics. Here we develop two complementary ones. The first is femtosecond stereotactic laser ablation, which rapidly creates complex cuts of subcellular structures and enables precise dissection of when, where and in what direction forces are generated. The second is an assessment of subcellular fluid flows by comparison of direct flow measurements using microinjected fluorescent nanodiamonds with large-scale fluid-structure simulations of different force transduction models. We apply these techniques to study spindle and centrosome positioning in early Caenorhabditis elegans embryos and to probe the contributions of microtubule pushing, cytoplasmic pulling and cortical pulling upon centrosomal microtubules. Based on our results, we construct a biophysical model to explain the dynamics of centrosomes. We demonstrate that cortical pulling forces provide a general explanation for many behaviours mediated by centrosomes, including pronuclear migration and centration, rotation, metaphase spindle positioning, asymmetric spindle elongation and spindle oscillations. This work establishes methodologies for disentangling the forces responsible for cell biological phenomena.

We present a new version of the fast Gauss transform (FGT) for discrete and continuous sources. Classical Hermite expansions are avoided entirely, making use only of the plane-wave representation of the Gaussian kernel and a new hierarchical merging scheme. For continuous source distributions sampled on adaptive tensor-product grids, we exploit the separable structure of the Gaussian kernel to accelerate the computation. For discrete sources, the scheme relies on the nonuniform fast Fourier transform (NUFFT) to construct near field plane wave representations. The scheme has been implemented for either free-space or periodic boundary conditions. In many regimes, the speed is comparable to or better than that of the conventional FFT in work per gridpoint, despite being fully adaptive.

We propose a practical finite volume method on cut cells using state redistribution. Our algorithm is provably monotone, total variation diminishing, and GKS (Gustafsson, Kreiss, Sundström) stable in many situations, and shuts off continuously as the cut cell size approaches a target value. Our analysis reveals why original state redistribution works so well: it results in a monotone scheme for most configurations, though at times subject to a slightly smaller CFL condition. Our analysis also explains why a premerging step is beneficial. We show computational experiments in two and three dimensions.

The high efficiency of a recently proposed method for computing with Gaussian processes relies on expanding a (translationally invariant) covariance kernel into complex exponentials, with frequencies lying on a Cartesian equispaced grid. Here we provide rigorous error bounds for this approximation for two popular kernels—Matérn and squared exponential—in terms of the grid spacing and size. The kernel error bounds are uniform over a hypercube centered at the origin. Our tools include a split into aliasing and truncation errors, and bounds on sums of Gaussians or modified Bessel functions over various lattices. For the Matérn case, motivated by numerical study, we conjecture a stronger Frobenius-norm bound on the covariance matrix error for randomly-distributed data points. Lastly, we prove bounds on, and study numerically, the ill-conditioning of the linear systems arising in such regression problems.

We numerically investigate the hydrodynamics and membrane dynamics of a multicomponent vesicle in two strongly confined geometries. This serves as a simplified model for red blood cells undergoing large deformations while traversing narrow constrictions. We propose a new parameterization for the bending modulus that remains positive for all lipid phase parameter values. For a multicomponent vesicle passing through a stenosis, we establish connections between various properties: lipid phase coarsening, size and flow profile of the lubrication layers, excess pressure, and the tank-treading velocity of the membrane. For a multicomponent vesicle passing through a contracting channel, we find that the lipid always phase separates so that the vesicle is stiffer in the front as it passes through the constriction. For both cases of confinement we find that lipid coarsening is arrested under strong confinement, and resumes at a high rate upon relief from extreme confinement. The results may be useful for efficient sorting lipid domains using microfluidic flows by controlled release of vesicles passing through strong confinement.

Substantial efforts have exploited reinforcement learning (RL) in the development of micro-robotic locomotion. These RL-powered micro-robots are capable of learning a locomotory policy based on their experience interacting with the surroundings, without requiring prior knowledge on the physics of locomotion in that environment. However, in their applications, micro-robots often encounter changes in the environment and need to adapt their locomotory gaits like living organisms in order to achieve robust locomotion performance. In standard RL methods, such a non-stationary environment can cause the micro-robots to continuously relearn the policy from scratch, degrading their locomotion performance. In this work, we explore a first use of a recently developed context detection method combined with deep RL to facilitate micro-robotic locomotion in a dynamically changing environment. As a proof-of-principle, we consider a simple micro-robot immersed in non-stationary environments switching between a viscous fluid environment and a dry frictional environment. We show that the RL with context detection approach enables the micro-robot to effectively detect changes in the environment and deploy specialized locomotory gaits for different environments accordingly to achieve significantly improved locomotion. Our results suggest the integration of deep RL with context detection as a potential tool for robust micro-robotic locomotion across different environments.

We introduce an electron-photon exchange-correlation functional for quantum electrodynamical density-functional theory (QEDFT). The approach, photon MBD (pMBD), is inspired by the many-body dispersion (MBD) method for weak intermolecular interactions, which is generalized to include both electronic and photonic (electromagnetic) degrees of freedom on the same footing. We demonstrate that pMBD accurately captures effects that arise in the context of strong light-matter interactions, such as anisotropic electron-photon interactions, beyond single-photon effects, and cavity modulated van der Waals interactions. Moreover, we show that pMBD is computationally efficient and allows simulations of large complex systems coupled to optical cavities.

mbedding materials in optical cavities has emerged as an intriguing perspective for controlling quantum materials, but a key challenge lies in measuring properties of the embedded matter. Here, we propose a framework for probing strongly correlated cavity-embedded materials through direct measurements of cavity photons. We derive general relations between photon and matter observables inside the cavity, and show how these can be measured via the emitted photons. As an example, we demonstrate how the entanglement phase transition of an embedded H

We explore the possibility to implement random walks in the manifold of Hartree-Fock-Bogoliubov wave functions. The goal is to extend state-of-the-art quantum Monte Carlo approaches, in particular the constrained-path auxiliary-field quantum Monte Carlo technique, to systems where finite pairing order parameters or complex pairing mechanisms, e.g., Fulde-Ferrell-Larkin-Ovchinnikov (FFLO) pairing or triplet pairing, may be expected. Leveraging the flexibility to define a vacuum state tailored to the physical problem, we discuss a method to use imaginary-time evolution of Hartree-Fock-Bogoliubov states to compute ground state correlations, extending beyond situations spanned by current formalisms. Illustrative examples are provided.

Neural quantum state (NQS) ansätze have shown promise in variational Monte Carlo algorithms by their theoretical capability of representing any quantum state. However, the reason behind the practical improvement in their performance with an increase in the number of parameters is not fully understood. In this work, we systematically study the efficiency of a shallow neural network to represent the ground states in different phases of the spin-1 bilinear-biquadratic chain, as the number of parameters increases. We train our ansatz by a supervised learning procedure, minimizing the infidelity w.r.t. the exact ground state. We observe that the accuracy of our ansatz improves with the network width in most cases, and eventually saturates. We demonstrate that this can be explained by looking at the spectrum of the quantum geometric tensor (QGT), particularly its rank. By introducing an appropriate indicator, we establish that the QGT rank provides a useful diagnostic for the practical representation power of an NQS ansatz.