Browse Articles
Discover research articles across all indexed journals
Investigation of tidally vigilant spatiotemporal foraging strategies of a Dugong (Dugong dugon) population in the macro-tidal Gulf of Kachchh, India
Near-atomic in situ architecture and membrane-coupled dynamics of the Vibrio cholerae sheathed flagellum
AI-driven Linear Diophantine Fuzzy evaluation for advancing quality and innovation in physical education
Broadband encoding and high-speed probabilistic bit generation with integrated microwave neurons
Multidimensional landscape persistence enhances the structural resilience of ecological networks in high-density urban agglomerations: a case study of the GBA
Abstract In high-density urban agglomerations, the conflict between rapid land expansion and ecological integrity poses a critical threat to regional sustainability. Existing ecological network planning mostly focuses on static connectivity based on single-temporal data, overlooking the stability of landscape units over long time series and their structural robustness against future uncertainties. To address this, taking the Guangdong-Hong Kong-Macao Greater Bay Area (GBA) (2000–2020) as a case study, this paper proposes an optimization framework integrating "MSP and Conflict Risk Rectification" (MSP-FT). First, Structural Persistence (SPI) and Functional Persistence (FPI) were quantified using long-time-series data. Second, a "Dual-threshold integration" strategy was implemented: a dual-threshold screening mechanism integrating “connectivity skeletons” and "high-quality cores" was adopted for source identification, and a spatiotemporal conflict risk penalty mechanism was introduced for resistance surface reconstruction. Finally, network resilience and effectiveness were evaluated by combining complex network theory and a "Topography-based Conservation Baseline (TCB)" benchmark. The results showed that: (1) The persistence strategy optimized network structure; compared with the traditional Static Ecological Network (SEN), the cumulative resistance cost of Persistence Ecological Networks (PENs) decreased by 30%, and the α index (closure) increased by 14.3%, forming a complex “cyclic redundancy” structure. (2) Robustness simulations confirmed that PENs maintained higher connectivity efficiency under intentional attacks due to their multi-center skeleton. (3) Crucially, overlay analysis against the TCB revealed a significant “Spatial Mismatch”: the current topography-dependent protection paradigm fails to cover approximately 54% of low-altitude source gaps and 88% of cross-regional corridor gaps. Accordingly, a hierarchical Redline-extension strategy was proposed: implementing "Remediation of Baseline Integrity" for source gaps and "Functional Connectivity Control" for corridor gaps. This study provides a scientific basis for constructing more adaptive ecological security patterns in high-density urban agglomerations.
Surface code scaling on heavy-hex superconducting quantum processors
Spatially-adaptive dynamic dictionary learning for fine-grained remote sensing image segmentation
Mol-CADiff: text-conditional molecule generation via causality-aware autoregressive diffusion
Abstract The design of molecules with desired properties is a key challenge in drug discovery and materials science. Traditional methods rely on trial-and-error, while recent deep-learning approaches accelerate molecular generation. However, existing models struggle with generating molecules based on specific textual descriptions. We introduce Mol-CADiff, a diffusion-based framework that uses causal attention mechanisms for text-conditional molecular generation. Our approach explicitly models the causal relationship between textual prompts and molecular structures, overcoming limitations in existing methods. We enhance dependency modeling both within and across modalities, enabling precise control over the generation process. While primarily designed for text-guided tasks, this architecture inherently supports unconditional generation, providing the added capability to autonomously sample the broader chemical space without explicit constraints. Here we show that Mol-CADiff outperforms alternative methods in generating diverse, chemically valid molecules, with better alignment to specified properties, enabling more intuitive language-driven molecular design. By bridging these modalities, our framework provides a versatile method for drug discovery.
Common mistletoe effects on Scots pine growth in northern Poland: a case study
Abstract The spread of common mistletoe ( Viscum album L.) and its increasingly frequent occurrence in forest stands have become important research and economic issues. Increased mistletoe occurrence has been observed in many European countries, and its effects on infested stands include slower growth, reduced needle biomass, and deteriorated wood quality. Empirical data were collected from an 80-year-old Scots pine stand, where diameter at breast height and tree height were measured for 500 trees. At the same time, the degree of mistletoe infestation was assessed and trees were assigned to three groups: uninfested trees (p0), trees with up to 50% crown infestation (p1), and trees with more than 50% crown infestation (p2). Mistletoe infestation was observed in 36.4% of all trees. From each infestation group, two sample trees were selected from each of five diameter classes using Urich’s method, and increment cores were collected at a height of 1.3 m using a Pressler increment borer. Annual radial increment for the period 2000–2022 was determined for each core using WinDendro software. Statistical analyses comparing p0, p1, and p2 trees, including ANOVA and the Kruskal–Wallis test, revealed statistically significant differences in radial increment in the final year of the study (2022). Pairwise comparisons (Student’s t-test, Welch’s t-test, and the Mann–Whitney U test) indicated statistically significant differences in annual radial increment at breast height between p0 and p2 trees over the last three years of the study (2020–2022). The strongest effect was observed in 2022, when the mean annual radial increment of trees with crown infestation exceeding 50% (p2) was more than 60% lower than that of uninfested trees (p0). These results indicate that the mere presence of mistletoe does not necessarily limit tree growth; rather, growth reduction depends on the degree of crown infestation. A marked reduction in radial increment occurs mainly at high infestation levels and is likely exacerbated by environmental stress.
Multi-omic analysis of androgen-deprived human testes reveals Sertoli cell regulation and niche reprogramming
Association between triglyceride glucose-waist height ratio index and incidence of type 2 diabetes in patients with nonalcoholic fatty liver disease
Chiral polariton transport enabled by optical spin Hall effect in perovskite waveguides
Trajectories and predictors of health-related quality of life among breast cancer survivors: a latent profile and transition analysis
Hybrid bioelectrochemical process enables hierarchical C, N, and P utilization towards negative carbon emission wastewater treatment
Synergistic effects of ceramic powder and nano-silica as supplementary cementitious materials in concrete
Fast-forward prediction of lattice Boltzmann dynamics with physics-informed neural operators
Abstract The lattice Boltzmann equation (LBE), rooted in kinetic theory, captures complex flow behaviour by evolving single-particle distribution functions (PDFs), but its explicit time-stepping makes large-scale simulation computationally intensive. Here we introduce a physics-informed neural operator framework that predicts the LBE evolution over large time jumps without performing step-by-step forward integration, bypassing the need to solve the collision kernel explicitly. The model embeds intrinsic moment-matching constraints and global equivariance of the distribution field, preserving the kinetic structure of the underlying system. The framework is discretization-invariant: models trained on coarse-grained PDFs perform inference on finer grids even when the relaxation time differs between resolutions. It is also agnostic to the lattice Boltzmann formulation, allowing the same architecture to be reused across different kinetic datasets. Across von Kármán vortex shedding, ligament breakup, and bubble adhesion, the framework offers a robust data-driven pathway for accelerating the lattice Boltzmann based dynamical systems.