arXiv subjects

Anton F. Burnet

Publications and source records attributed to Anton F. Burnet.

4 recordsLinked to original sources

arXiv2026

State-dependent recruitment of adhesion molecules enables perfect stabilization in cell-adhesion models

Cells adapt their adhesion to mechanical load, but the physical conditions under which this response prevents rupture remain unclear. Inspired by focal adhesions, we study models in which force-sensitive conformational states within an adhesion cluster are coupled to the recruitment of additional molecules. We show that sufficiently strong coupling creates a distinct regime under load in which clusters grow in proportion to the applied force while the average load per bond remains below the level that destabilizes the cluster. The clusters can therefore withstand arbitrarily large stationary forces in principle. We term this behavior ``perfect stabilization''. At still stronger coupling, the same feedback causes unbounded growth already at equilibrium. For the minimal model, we derive state diagrams that characterize adhesion stability under stationary and dynamic loading. More broadly, using generic molecular-state networks, we derive conditions under which perfect stabilization and related growth instabilities arise in nonequilibrium adhesion systems with state-dependent recruitment.

cond-mat.soft↗

arXiv2026

Energetics of stochastic limit-cycle oscillators: when does coupling reduce dissipation?

Non-linear oscillators serve important functions in many biological systems, including within the inner ear and neuronal networks. The sustainment of oscillations in noisy environments requires continuous energy dissipation, quantified by the steady-state entropy production rate (EPR). We study an idealized, analytically tractable model of a stochastic circular limit cycle and examine how mutual coupling in pairs and populations alters dissipation. For a single oscillator, the EPR depends on three key factors: intrinsic frequency, tangential velocity fluctuations, and mean tangential velocity. The dynamics are characterized by a dimensionless effective temperature given by the ratio of intrinsic relaxation and diffusion timescales. For radial (amplitude), phase (Kuramoto-like), and Cartesian couplings, we derive analytical expressions for the EPR and confirm them numerically. Varying the effective temperature and system size strongly influences how the EPR depends on coupling strength and, in some cases, results in qualitatively distinct behaviors. Moreover, the coupling types affect the tangential velocity distributions differently. Notably, in all cases studied, Cartesian coupling reduces the EPR relative to the uncoupled system, irrespective of effective temperature and system size. The analysis of idealized non-linear oscillators reveals that different classes of coupling interactions and competing timescales present in the oscillators have distinct effects on energy dissipation.

cond-mat.stat-mech↗

arXiv2026

Diffusion of gravitactic chiral active Brownian particles in an asymmetric channel

The diffusion of micro- and nanoswimmers in a fluid, confined within irregular structures that impose entropic barriers, is often modeled using overdamped active Brownian dynamics, where viscous effects are paramount and inertia is negligible. Here, we numerically investigate the diffusive behavior of chiral self-propelled particles in a two-dimensional asymmetric channel subjected to an external torque arising from a gravitational field. We reveal the emergence of resonant diffusion when the external torque $ω$ approaches the intrinsic angular velocity $ω_{0}$ of particles. This resonance manifests as a pronounced accumulation of particles near the upper-left corner of the channel, accompanied by an enhanced peak in the effective diffusion coefficient. In particular, it is observed only for low rotational diffusion rates and does not persist beyond moderate values of $ω_{0}$. Prominent transport features, such as rectification at low values of $ω$, a monotonic increase in average velocity with $ω$, and a nonmonotonic response of transport characteristics (average velocity and effective diffusion coefficient) as a function of the rotational diffusion rate near the resonance point, are explained. Furthermore, we show that the transport characteristics depend strongly on the aspect ratio of the channel. For instance, the enhanced diffusion peak becomes more pronounced with increasing aspect ratio, and the average velocity saturates at higher values for wider bottleneck openings. It is conceivable that these findings have a great potential for developing microfluidic and laboratory-on-a-chip devices for particle separation, targeted drug delivery, and advanced active materials.

cond-mat.soft↗

arXiv2025

Predicting solvation free energies with an implicit solvent machine learning potential

Machine learning (ML) potentials are a powerful tool in molecular modeling, enabling ab initio accuracy for comparably small computational costs. Nevertheless, all-atom simulations employing best-performing graph neural network architectures are still too expensive for applications requiring extensive sampling, such as free energy computations. Implicit solvent models could provide the necessary speed-up due to reduced degrees of freedom and faster dynamics. Here, we introduce a Solvation Free Energy Path Reweighting (ReSolv) framework to parametrize an implicit solvent ML potential for small organic molecules that accurately predicts the hydration free energy, an essential parameter in drug design and pollutant modeling. With a combination of top-down (experimental hydration free energy data) and bottom-up (ab initio data of molecules in a vacuum) learning, ReSolv bypasses the need for intractable ab initio data of molecules in explicit bulk solvent and does not have to resort to less accurate data-generating models. On the FreeSolv dataset, ReSolv achieves a mean absolute error close to average experimental uncertainty, significantly outperforming standard explicit solvent force fields. Compared to the explicit solvent ML potential, ReSolv offers a computational speedup of four orders of magnitude and attains closer agreement with experiments. The presented framework paves the way toward deep molecular models that are more accurate yet computationally cheaper than classical atomistic models.

physics.chem-ph↗