Browse Articles

Genetic diversity and recombination of begomoviruses associated with chilli leaf curl disease in northern and Central-East India

Early intraocular pressure dynamics after intravitreal aflibercept 8 mg versus 2 mg

Turkish nurses’ attitudes towards violence against women, level of recognition of signs, and intention to report violence against women: a cross-sectional study

Factors associated with meconium stained amniotic fluid and birth outcomes among mothers at Woldia Comprehensive Specialized Hospital

Levels and determinants of healthcare-seeking delay for pediatric pneumonia: a mixed-methods study

Image-based multiphysics modelling of random metallic meshes

Abstract Bio-inspired conductive networks are increasingly exploited in transparent heaters and smart multifunctional structures due to their inherent scalability, robustness, and efficient transport properties, yet their intrinsic disorder poses major challenges for quantitative modelling. Here we introduce a computationally efficient, image-driven framework for the automated digitisation, reconstruction, and multiphysics analysis of random conductive meshes. High-resolution microscopy images are converted into graph-based network representations via skeletonization, branch tracking, and connectivity refinement, preserving local geometry while drastically reducing degrees of freedom. An analytical model is developed to predict sheet resistance directly from extracted network metrics, enabling rapid, non-destructive electrical characterisation. The digitised networks are further used to perform coupled electro-thermal and electro-thermo-mechanical finite-element simulations. Experimental validation using silver mesh heaters embedded in polymer laminates shows excellent agreement with predicted electrical resistance, temperature evolution, and stiffness modulation under electrical loading. The framework reduces analysis time and enables systematic assessment of structural features, such as dangling branches, on device performance. This approach provides a scalable route for predictive design and optimisation of fractal-based multifunctional electronic materials.

Enhancing the fertilizing properties of digestate’s solid fraction by Trichoderma-driven microbial activation: In-field application effects on tomato growth

Exosomal PD-L1 level in plasma as a predictive marker for prognostic and clinical correlation in NSCLC

A novel pray optimization algorithm for six degree of freedom robotic arm trajectory planning

Abstract This study introduces the Pray Optimization Algorithm (POA), a novel metaheuristic inspired by the procedural rituals of Islamic pray, designed to solve complex engineering and robotic manipulator problems. The mathematical model is structured into three distinct phases: Phase I simulates searching for a suitable mosque; Phase II models congregational alignment (lining up for pray); and Phase III implements a cumulative scoring system to drive convergence. The efficacy of the POA is initially validated on the CEC2017 benchmark suite through comparative analysis with six high-performing metaheuristic algorithms. The algorithm’s statistical superiority is subsequently confirmed using the Wilcoxon rank-sum and Friedman tests. To verify its practical applicability, the POA is deployed to solve three classical real-world engineering optimization problems. Ultimately, the algorithm demonstrates superior performance when compared against five state-of-the-art methods in optimizing the trajectory planning of a 6-DOF industrial robotic arm. These findings substantiate the effectiveness of the proposed POA in navigating constrained, real-world engineering search spaces.

Impact of currents on non-equilibrium coexistence in chemically driven mixtures

Virtually every biological function emerges through the organization of molecules in time and space. Consequently, a major challenge in statistical physics is to uncover the universal principles governing macromolecular self-organization within the crowded, non-equilibrium environment of the cell. Here, we investigate a class of models where molecules maintain a conserved total concentration but can switch “identities,” thereby modulating their intermolecular interactions. By enforcing thermodynamic consistency via the local detailed balance condition, we derive the steady-state criteria determining coexisting concentrations in a binary mixture. In particular, local detailed balance together with state-dependent kinetics impose a difference in chemical potentials across the interface, and we obtain jump conditions that generalize Gibbs’ equilibrium coexistence criteria of equal pressure and chemical potential to chemically driven steady states. Maintaining the chemical potential differences requires particle currents, which are confined to the interfacial region.

RMSD: an interpretable framework for streaming-compatible multi-source data fusion and skill-gap diagnosis in vocational education

Abstract Vocational education often faces a mismatch between curriculum design and evolving enterprise skill demands. This paper proposes RMSD, an interpretable framework for streaming-compatible multi-source data fusion, student–job matching, and skill-gap diagnosis. RMSD integrates student records, behavioral logs, project and internship information, job descriptions, and enterprise feedback. It combines BERT-based semantic encoding, temporal self-attention, CFN-based multimodal fusion, and hierarchical skill-tree context to estimate matching scores and node-level skill gaps. Experiments on student–enterprise interaction data show that RMSD outperforms representative baselines, including DeepFM, NCF, BERT-Dual Encoder, Transformer-Seq, and Skill-KG Matching. Compared with DeepFM, RMSD improves HR@5 and MRR@10 by 5.7 and 4.7 percentage points, respectively. It also reduces the average skill-gap score from 0.42 to 0.30 during the observed curriculum-intervention period. System-level evaluation shows that RMSD supports near-real-time updates under the tested institutional workload. These results suggest that RMSD offers a practical and interpretable approach for data-driven vocational education analytics.

Linear active polymers in porous media: The type of activity matters

Active polymers represent a distinct class of non-equilibrium systems whose structure and movement are strongly influenced by the nature of activity and environmental complexity. In this work, we present a comparative study of an Active Brownian Polymer Chain (ABPC) and a Tangentially Active Polymer Chain (TAPC) in a two-dimensional porous environment using Brownian dynamics simulations. Our study on two different types of active polymers reveals distinct differences in their conformational and dynamical properties. We identify fundamentally different trapping mechanisms in the two systems: ABPC exhibits extensive folding in the porous confinement, whereas TAPC adapts stable spiral conformations accounting for persistent self-trapping at moderate and high activity. Furthermore, ABPC undergoes unrestraint motion at high activity, while TAPC displays reptation-like motion at low activity, leading to efficient escape from the traps. Our findings suggest that the nature and strength of activity coupled with environment complexity lead to emergent phenomena that dictate these unique conformational and dynamical properties of active polymers.

Comprehensive characterization of lithium carbonates recovered from different spent lithium-ion batteries black masses

Abstract The growing demand for electric vehicles has intensified the need for efficient recycling of lithium-ion batteries (LIBs) to recover critical materials. This work investigates the recovery and detailed characterization of Li 2 CO 3 samples recovered from two different types of black masses (BM) derived from spent NMC 622 batteries. Structural, morphological, and compositional analyses were carried out using X-ray Diffraction (XRD), Scanning Electron Microscopy (SEM), Raman Spectroscopy (RS), and X-ray Microanalysis (EDX). The obtained results reveal that pretreatment of the starting black mass significantly influences the impurity content and crystalline structure of the recovered samples. The sample from thermally-pretreated BM exhibited higher purity and fewer secondary phases than those from mechanical-pretreated BM, where compounds such as Li 2 SO 4 , LiAl 2 (OH) 6 ·H 2 O, LiF, and LiOH were detected. Morphological differences correlated with the presence of impurities impacting crystal growth. These findings provide valuable insights for optimizing pretreatment and purification routes, thereby improving the quality of recycled Li 2 CO 3 for battery manufacturing and supporting circular economy strategies in the LIB sector.

TorchFF: A high-performance GPU-accelerated differentiable force field library

Molecular dynamics (MD) and the development of next-generation force fields increasingly rely on automatic differentiation for efficient simulated property prediction and parameter optimization. However, standard FFs and machine learning frameworks often suffer from significant performance bottlenecks—such as kernel launch overhead and memory bandwidth limitations—when executing the many-atom, small-kernel operations characteristic of MD simulations. Here, we present TorchFF, a high-performance, differentiable library that extends PyTorch with a suite of customized CUDA operators specifically engineered for molecular modeling. By implementing performance-critical routines—including bonded interactions, multipolar electrostatics, particle mesh Ewald, and neighbor list searches—as backend-optimized primitives, TorchFF bridges the gap between the flexible Python ecosystem and the execution speed of compiled MD engines.

AI‑enabled resource management for 6G‑IoT hybrid systems: a unified simulation platform

Abstract The integration of Artificial Intelligence (AI) with sixth‑generation (6G) communication technologies is expected to transform resource management in Internet of Things (IoT) systems, where reliability, latency, and adaptability are critical. However, hybrid 6G–IoT environments combine two fundamentally different subsystems, resource‑constrained IoT devices and ultra‑high‑performance 6G infrastructure, creating a highly complex operational space with a multitude of interacting parameters. This results in severe heterogeneity across frequency bands, latency requirements, traffic behaviors, and computational capabilities. Such heterogeneity makes end‑to‑end modeling, resource management, and optimization extremely challenging when capturing the dynamics of both IoT endpoints and 6G networks. To address these challenges, this work presents 6G‑IoT‑Sim, a modular 6G‑enabled IoT simulator that provides a unified and extensible platform for implementing and analyzing 6G‑IoT networks. The platform incorporates 6G architectural capabilities, a configurable network‑design interface, a diverse dataset generator tool, and a real‑time monitoring dashboard. Additionally, it integrates a hybrid AI-based framework comprising different AI models and optimization techniques for intelligent resource management. A multidimensional performance evaluation demonstrates the effectiveness of the framework for intelligent resource management across diverse 6G‑IoT conditions. The results further highlight the potential of the proposed simulator as a flexible and extendable platform for advancing adaptive solutions in 6G‑enabled IoT systems.

Self-phoretic oscillatory motion in a one-dimensional channel

We study a simple model for a particle that is active due to repulsive self-phoresis and that has been proposed to model symmetric camphor grains. The particle generates a concentration field through the continuous emission of a chemical substance and its motion is driven by gradients of this field as it diffuses within a confined channel whose ends perfectly reflect the chemical. The reflection of the chemical field leads to an effective confinement of the particle, which itself is reflected before encountering the channel ends. The system displays a transition from a passive state, where the particle rests at the channel midpoint, to an active state characterized by highly regular, non-chaotic oscillations. We analytically construct the phase diagram and derive the oscillation frequency and amplitude in the vicinity of the transition. A perturbative analysis perfectly describes the dynamics of the particle, even for oscillations as large as half the channel size. Furthermore, we develop an analysis which explains the mechanism of particle reflection close to the channel edges in the regime of large activity.

Trends and predictors of pediatric cardiovascular mortality: a 2014–2024 multicenter retrospective cohort study in Hawassa, Ethiopia

Limits of the non-linear generalized Langevin equation: Cross-correlations, irreversibility, and desynchronization

The generalized Langevin equation (GLE) is widely used to model complex soft-matter systems, including biomolecular dynamics, by incorporating memory effects and colored noise into coarse-grained descriptions. However, recent results suggest that combining memory with non-linear forces, which are ubiquitous in soft matter, introduces fundamental analytical inconsistencies. Here, using a simplified model, we investigate the practical numerical consequences of these analytical results in equilibrium systems. We show that non-linear forces generate cross-correlations with the noise, modifying the fluctuation–dissipation theorem and rendering the noise position-dependent and irreversible. This implies that the commonly assumed reversible Gaussian noise in GLE simulations fails to capture essential features of the microscopic fluctuations. For weak non-linearities, these issues can be partially resolved either by using an iterative optimization of memory or by using microscopically consistent noise, which unexpectedly synchronizes GLE trajectories with the underlying microscopic dynamics. For stronger non-linearities, such as high barriers or shoulders in the external potential, however, iterative reconstruction fails and we observe desynchronization, indicating that the non-linear GLE no longer correctly reproduces the microscopic dynamics. Our results show in which situations non-linear GLEs can be accurately applied and when they fail, thus providing practical guidance for their application to coarse-grain soft-matter systems.

Yield stability of groundnut genotypes across contrasting environments in Niger using GGE biplot analysis

Coherent dynamics of the spin-boson model in the ultra-strong coupling regime

Quantum mechanics describes the unitary time evolution of closed systems. In practice, every quantum system interacts with the environment, leading to an irreversible loss of coherence. The spin-boson model (SBM) is central to the understanding of the fundamental process of decoherence of a two-state quantum system interacting with a bosonic heat bath, but the nature of transient dynamics in the presence of hybrid diagonal and off-diagonal system–bath interactions remains much less explored. Here, we investigate how the hybrid system–bath interactions of an Ohmic environment induce localization in the bias-free SBM. For strong coupling to the environment, localization is strongly affected by a dynamically generated bias via the renormalization of the tunneling amplitude. We find that counteracting effects of Hamiltonian parameters on non-exponential short-time dynamics and long-time population equilibration can lead to a separation of timescales and transient quantum coherent dynamics that can persist even for ultra-strong system–bath interaction. The findings offer novel insight into the equilibration behavior of quantum devices operating in the ultra-strong coupling regime.