Cyclic peptides targeting FXR show promise in treating MASH by reducing intestinal ceramide production and promoting GLP-1 release. DC646 is the most potent FXR antagonist with a favorable safety profile.
Researchers developed a novel approach to address thermal challenges in electric vehicle charging using a gallium-based liquid metal flexible charging connector. The LMFCC demonstrated excellent flexibility, high transmission stability, and efficient heat dissipation.
Current methods like static cold storage and machine perfusion have limitations, with only 10% of global demand met. Researchers explore vitrification cryopreservation as a promising alternative to address toxicity issues and improve transplant survival rates.
Crop domestication and improvement significantly reshape root traits, including lateral root density, main root length, and cortical cell size. The structure and function of the associated microbial community also undergo profound changes, affecting nutrient absorption and immune capacity.
The study explores innovative mixing methods for overcoming traditional limitations in chemical reactors. Researchers developed models and simulations to simulate macromixing and micromixing, leading to improved industrial production efficiency and product selectivity.
Researchers propose a standardized taxonomy of cascading hazards and interactions between people and urban systems. An integrated risk assessment framework considers multiple dimensions and individual behaviors to improve early warning systems.
Climate change and air pollution threaten food systems, while agricultural activities contribute to global warming and poor air quality. Implementing mitigation and adaptation strategies within the food system can reduce emissions and improve resilience.
Researchers have made significant progress in applying tissue engineering to spinal cord injury (SCI) repair. Biomaterials such as hydrogels and decellularized extracellular matrix promote nerve regeneration, while stem cells and exosomes enhance functional recovery.
A recent study highlights both positive and negative effects of elevated CO2 on plants, affecting nutrient assimilation, growth, and crop quality. Crop varieties with superior biofortification traits can help alleviate nutrient deficiency.
Research reveals that environmental nutrient conditions and plant genes regulate root microbiota composition, while the microbiota improves nutrient availability and regulates plant hormones. This mutualistic relationship enhances plant growth and development.
The SSA model leverages semantic structure information to generate high-quality CAMs without parameters in the model inference step. It achieves new state-of-the-art performance on both WSOL and WSSS tasks, demonstrating its potential for improving weakly supervised pixel-wise dense predictions.
Covert communication, a technique to hide communication behavior, offers an extra layer of protection. Researchers explored various domains and identified future research trends, including intelligent cooperative and parasitic methods.
Researchers propose future data collection and processing principles to address issues in data accumulation, modeling, and application. A global standard protocol framework is needed for a high-quality data ecosystem and healthy AI development.
A recent study uses deep reinforcement learning to tackle complex manufacturing scheduling problems. The proposed IPPO method outperforms other approaches in terms of convergence and solution quality, achieving better invert generational distance and purity values.
Researchers have made significant advancements in intracortical neural interfaces for freely moving animals, enabling better understanding of neural circuit functions and mechanisms. These technologies also hold promise for developing more effective therapies for neurological diseases.
Researchers explore three key directions to address LLM limitations: knowledge empowerment, model collaboration, and co-evolution. Techniques like retrieval-augmented inference and knowledge prompting enhance factual accuracy and interpretability.
Researchers developed a novel core-shell structured composite to enhance polypropylene's performance at low temperatures. The addition of HDPE improved impact strength and adjusted brittle-ductile transition temperature, while maintaining tensile strength.
LearningEMS provides a unified platform for various EV configurations, supporting detailed comparisons of multiple algorithms and benchmarks. The study highlights the strengths and weaknesses of different approaches, emphasizing the importance of state, reward, and action settings in RL-based EMS optimization.
Researchers have developed an innovative acoustofluidics-based approach for delivering biomolecular cargos into cells, overcoming limitations of traditional methods. The new method enables controlled contact between cells and nanoparticles, enhancing membrane permeability and improving delivery efficiency.
The NDTSL method uses Schatten-p norm and soft label matrix to improve model classification performance on target data. It effectively tackles the problem of label noise in source labels, achieving better results than previous research.
Researchers developed oncomicrobial vaccines targeting C. jejuni and ETBF, which reduced tumor development and inflammation in mice models. The vaccines elicited strong immune responses while minimizing disruption to gut microbiota composition.
A systematic re-evaluation of China's Ordovician tectonic architecture proposes a refined model of four mega-regions and 16 sub-regions, resolving decades of ambiguity. The revised framework identifies key cratons and microcontinents as mobile crustal fragments within ancient oceans.
Researchers from Professor Jing Zhang and Professor Yueguo Chen provide a comprehensive review of recent academic research on Large Language Models (LLMs) for table processing. The study highlights the limitations of current AI methods in addressing complex table processing tasks and ambiguous user queries.
A landmark study resolves debates on mantle dynamics beneath the Tonga Subduction Zone by mapping three-dimensional shear-wave velocity and azimuthal anisotropy structure. The research team analyzed 150,219 amplitude and phase measurements to construct a high-resolution 3D velocity model extending to 300 km depth.
Researchers applied scaling laws to urban ecosystems, discovering multistable states and threshold effects that impact ecological services. Optimizing green spaces and sustainable infrastructure can enhance urban ecosystem resilience, ultimately improving quality of life for residents.
Researchers developed uPEn to enhance genome stability and repair efficiency, achieving efficient Kozak motif insertion and precise modifications in mouse and sheep zygotes. The platform showed potential for complex genetic modifications, disease modeling, and gene therapy applications.
A new model based on Dual-view Prompt and Element Correlation (DPEC) encodes spatial discriminative information to enhance spatial relation extraction. The approach combines a mask language model with two prompts: Link Prompt for contextual information and Confidence Prompt for trigger recognition.
A novel method using sodium salts of FAs, ethanol, and ultrasonication creates stable FA micelles with no degradation of FAs during preparation. This approach preserves the integrity of FAs and is applicable to a wide range of fatty acids.
Gypsum and silica scaling significantly impact membrane desalination performance, causing rapid kinetics and adhesive layers. Researchers develop targeted strategies to combat mineral scaling using hydrophilic polymer brushes, zwitterionic coatings, and modified membrane surface charges.
Researchers found that combining Rhizobium inoculation with low phosphorus application increased plant growth, root nodulation, and grain yield of common bean (Phaseolus vulgaris) in Ethiopia. The study suggests recommending this agronomic input for resource-poor farmers to increase local common bean yield.
Researchers developed a groundbreaking pipeline for rapid pathogen detection by automating primer design across entire genomes. This innovation enhances speed and accuracy in diagnosing infectious diseases, particularly for viruses and fungi.
A new method using satellite observations of NO2 distinguishes anthropogenic emissions from natural processes, providing precise way to identify emissions from point sources. The study offers a scalable solution for tracking emissions, from localized sources to broader national levels.
A virtual teaching and research section operated as an open source community has developed a course knowledge graph and implemented distinctive teaching methods. The model has incubated several national-level first-class courses and won teaching achievement awards.
Researchers developed a prognostic model for acute myeloid leukemia based on ferroptosis-related lncRNAs and immune infiltration analysis. The study found that this model holds promise for improving diagnosis, prognosis, and treatment strategies for AML patients.
A recent study analyzed the power response of networks to voltage with time-varying amplitude and frequency (TVAF), revealing time-varying active and reactive power. The research proposed a simplified calculation method for practical engineering applications.
The study proposes a digital pedagogy for sustainable educational transformation (DP4SET) framework that advocates for digital competence and evidence-based practice in the digital era. The DP4SET framework has been successfully applied in Chinese contexts, reshaping teaching and learning practices.
Recent research reviews how 3D printing enhances microbial electrochemical systems by improving design flexibility, precision, and bioprinting applications. This enables optimized reactor designs, tailored electrode fabrication, and improved system performance.
Researchers developed new framework to utilize higher-order neighbors for sparse KG completion, alleviating performance decrease caused by sparsity. The proposed method employs Reinforcement Learning and Markov Logic Network to mine reasoning paths and encode relations sequence.
A new PEC strategy converts PS waste into valuable products, offering a promising solution for sustainable plastic waste treatment. The researchers achieved efficient oxidative degradation of PS under sunlight illumination.
Multi-photon 3D nanoprinting enables 3D processing beyond the diffraction limit, with applications in optics, biology, and mechanical engineering. However, slow processing speed and materials limitations hinder its large-scale industrial production.
The study developed a framework guided by the SENCE theory, using pressure-state-response models and indicators like population density and net primary productivity. Ecological security improved significantly in the basin from 2000 to 2020, with higher levels in western and eastern regions.
A recent study found that electric vehicles can lead to increased commuting distances and times, while households with EVs tend to live farther from city centers due to lower operational costs. The study also highlights the role of factors like vehicle age, purchase price, and charging infrastructure in shaping urban development.
Researchers have made significant progress in nonlinear meta-devices, which can enhance nonlinear optical responses at the nanoscale. These devices can boost efficiency without phase-matching, leading to applications such as second-harmonic generation and harmonic imaging.
A new design paradigm for reconfigurable intelligent surfaces (RISs) offers a fresh approach to address challenges in RIS design. The paradigm combines a novel topological representation method and separate design architecture, reducing pattern solution space and dataset acquisition costs.
A recent study published in Engineering has shed light on a significant cybersecurity risk facing smart grids due to increasing complexity and integration of distributed power supplies. False data injection attacks (FDIAs) targeting data-driven algorithms can compromise system operations, reducing prediction accuracy by up to 41.75%.
The study reveals the molecular mechanisms underlying endothelial-to-osteoblast conversion, promoting osteoblast differentiation and influencing Runx2 expression. The Kindlin-2/Piezo1/TGFβ/Runx2 pathway contributes to bone tissue homeostasis and may have implications for treating bone diseases like osteoporosis.
A novel self-biased hybrid system incorporating spectral beam splitters enhances photoelectrochemical water splitting performance. The system boosts hydrogen production and solar-to-hydrogen efficiency compared to conventional systems.
Researchers found that phytosiderophores, such as proline-2'-deoxymugineic acid (PDMA), play a crucial role in regulating the rhizosphere dialogue between plants and microorganisms. PDMA enhances nutrient availability and promotes beneficial rhizobacterial community formation.
A new method for quantifying 3D cell shape during early embryogenesis, called 3DCSQ, has been developed to analyze cellular morphologies. This method combines spherical grids, spherical harmonics, and principal component analysis to identify biologically reproducible cellular patterns in organisms like C. elegans.
A recent study explores digital twin (DT) technology in road engineering, finding applications across all lifecycle phases. DT enables optimizing route selection, pavement design, resource allocation, and maintenance decisions, but faces challenges such as lack of standardization and innovation.