Xichang University

This study conducts acoustic emission(AE) tests on five typesof medium-grained sandstones under uniaxial compression coupled withanalysis using nuclear magnetic resonance (NMR) and scanning electronmicroscopy (SEM) techniques. We systematically investigated the relationshipamong mineral content, pore structure (porosity and pore type), mechanicalproperties, damage mechanisms, and AE characteristics. The resultsshow that the uniaxial compressive strength (UCS) and modulus of elasticity(E) of sandstones are positively correlated with the content of feldsparand other minerals, while exhibiting an inverse relationship withquartz content. Simultaneously, they are negatively correlated withthe porosity and influenced by the percentage of mesopores. The AEenergy and cumulative AE energy characteristics are jointly influencedby the pore structure and fracture mechanism. The correspondence betweenSEM fracture morphological features and the change rule of fractaldimension (D) was established based on fractal theory. Additionally,the crack classification results based on the ratio of rise time toamplitude (RA) and average frequency (AF) align with the macroscopicfailure mode. However, the cumulative number of cracks and their correspondingcumulative AE energy are jointly influenced by both the pore structureand the fracture mechanism. The distribution characteristics of theAE dominant frequency are influenced by porosity and correlated withthe complexity of internal fracture patterns. Furthermore, the “three-frequencysynergistic growth phenomenon” occurs during macroscopic damage.The research findings provide a scientific basis for the assessmentand monitoring of sandstone rock stability.

The increasing complexity of modern neural networks impedes systematic understanding and formal analysis,often resulting in “black box” behavior. Our paper introduces Typed Petri Nets (TPNs) as a formal framework formodeling the structure and dynamics of neural networks, aiming to address these challenges. We define the corecomponents of TPNs―places, transitions, arcs, a type system (e.g., for tensors), and computation assignments―anddemonstrate how they can compositionally represent basic network layers. Furthermore, we illustrate the capability ofTPNs to model advanced architectures, including convolutional layers, skip connections, and multi-branch structures.TPN dynamics naturally simulate forward propagation through markings and firing rules. The framework’sformal semantics can be rigorously explored using category theory, as illustrated by applications in probabilistic contextsusing Markov Categories. Key advantages of the TPN approach include modularity through compositionality,explicit information flow representation, type safety, and a foundation for formal analysis. This work establishesTPN as a structured, mathematically grounded perspective for understanding complex neural networks, offering apromising basis for future research into learning dynamics, verification, and tool development.

In response to the challenges posed by the large number of small targets, complex backgrounds and significant computational load involved in detecting olives, this study presents YOLO-TinyFuse, a lightweight detection model developed based on YOLOv8n. This model incorporates the P2 high-resolution feature layer, a ModifiedNeck cross-scale fusion structure (ModifiedNeck) and a bidirectional feature pyramid network (BiFPN) dynamic weighting module within a unified architecture. This architecture simultaneously preserves high-resolution feature representations, enhances bidirectional multi-scale interaction and optimises weighted feature aggregation. This synergistic design substantially improves the recognition of small objects while reducing model complexity further. Evaluations conducted on a multi-scenario olive phenotyping dataset demonstrate that YOLO-TinyFuse achieves an mAP50 of 92.3% and a Recall of 84.5%. This represents improvements of 2.6% and 3.2% respectively over YOLOv8n, while reducing the parameter count by 6.76%. These results confirm that the proposed model provides a deployable, computationally efficient, real-time solution for target recognition on mainstream edge computing platforms in automated olive harvesting scenarios, and offers a reusable, lightweight framework for agricultural small-object detection tasks requiring high performance and optimised computational efficiency.

Rotavirus is a leading cause of severe, dehydrating diarrhea in infants and young animals, causing significant global morbidity and mortality. For decades, research was hindered by challenges in establishing reverse genetics systems due to the virus's complex segmented genome and poor cell culture adaptation. The first helper virus-dependent system (2006) was limited by low efficiency. A 2017 breakthrough established the first fully plasmid-based system, which eliminated helper viruses and revolutionized the field. Subsequent optimizations, such as codon modification and CRISPR/Cas9 integration, have significantly enhanced efficiency, enabling viable systems for diverse human and animal strains. This narrative review summarizes the evolution and technological milestones of rotavirus reverse genetics. We discuss critical applications in analyzing viral gene function, developing novel vaccines, screening antiviral drugs, and investigating cross-species transmission. Finally, we provide an outlook on the future prospects of this transformative technology.

Carotenoids are a class of C40 isoprenoid-derived fat-soluble pigments that play vital roles in plant physiology and human health and serve as precursors for several biologically critical regulatory molecules. Carotenoid cleavage dioxygenases (CCDs) are key enzymes that catalyze the selective oxidative cleavage of carotenoids into apocarotenoids, thereby significantly influencing plant development and responses to abiotic stress. Although extensive research has been conducted on many model species, comprehensive studies on the StCCD gene family in potato remain limited. In this study, we conducted a genome-wide analysis to identify and characterize the CCD gene family in potato. Phylogenetic and structural analyses classified the 17 StCCD genes into six distinct subfamilies, which are distributed across five chromosomes of the genome. Analysis of cis-acting regulatory elements revealed that the promoters of most StCCD genes contain various elements associated with light responsiveness, stress signaling, and phytohormone regulation. Molecular docking analysis indicated that CCD proteins exhibit distinct substrate specificity in their binding to carotenoids and intermediate products. The expression profiling of StCCD genes uncovered pronounced specificity in their expression, which was evident across tissues, throughout tuber maturation, and following exposure to abiotic stresses and hormonal applications. This specificity strongly implicates these genes in the regulation of developmental processes and stress adaptation mechanisms. This study provides a comprehensive genomic and transcriptomic overview of the CCD gene family in potato, establishing a foundation for functional characterization of individual genes in the future.

Sustainable management of flue-cured tobacco requires a careful balance among productivity, chemical composition, and soil ecological function, which are often disrupted by excessive chemical fertilization. This study aimed to elucidate how microbial fertilization regulates plant performance, chemical coordination, and rhizosphere microbial structure under field conditions. A two-year factorial field experiment was conducted in Sichuan, China, using a Bacillus-based plant growth-promoting microorganism (PGPM) and a commercial microbial consortium (Xi⋅Weifeng), applied individually or in combination at gradient doses. Agronomic traits, cured-leaf chemical composition, secondary metabolites, and rhizosphere bacterial communities were comprehensively analyzed using multivariate statistics, network correlation analysis, and structural equation modeling (SEM). Moderate PGPM application (27 kg⋅ha–1) significantly increased plant height (8.6%), internode length (15.3%), and leaf width (7.8%) at the vigorous growth stage. Co-application further enhanced leaf expansion (9.7%) and improved chemical coordination, maintaining optimal sugar/nicotine (8–12) and N/nicotine (0.7–1.0) ratios. Chlorogenic acid (18.8 mg⋅g–1) and neochlorogenic acid (2.7 mg⋅g–1) were markedly elevated under the A27B54 treatment. Rhizosphere bacterial diversity peaked under co-application, with Bacillus, Rhizobiales, and Sphingomonas emerging as key taxa positively associated with both metabolic and agronomic improvements. SEM demonstrated that fertilization effects on leaf quality were mediated indirectly through microbial community restructuring and metabolite modulation. Microbial fertilizer co-application enhances tobacco performance by promoting rhizosphere microbial diversity and functional coordination, which in turn improves metabolic balance and nutrient-use efficiency. These findings highlight a soil microbiome-mediated pathway linking fertilization strategy to crop physiological and chemical responses, providing mechanistic insights for sustainable fertilization management.

Mesoscale convective systems (MCSs) over the Tibetan Plateau (TP) strongly influence downstream rainfall. In plateau regions, existing single-threshold identification algorithms often have high misidentification rates in plateau regions, because cold cirrus can be misclassified as convection when using only blackbody brightness temperature (TBB), and the complete evolution of MCSs is not fully represented by conventional area-overlap tracking. This study improves an area-overlap combined with Kalman-filtering tracking algorithm (AOL-KF) for the warm seasons (May–October) of 2019–2023 over the TP by introducing a rainfall constraint, in which FY-4A TBB is coupled with GMCP merged rainfall through a Rain-rate threshold (Rrt) and FY-4A cloud type (CLT) is incorporated as auxiliary cloud information. This study improves MCS identification by integrating GMCP precipitation data, and the rainfall constraint is further evaluated using Ka-band ground-based millimeter-wave cloud radar (GB-MMCR) observations at Naqu and Yushu during July–August 2020. The results show that cirrus-induced misidentification is suppressed, and the resulting satellite-based MCS identification and tracking are more consistent with radar diagnostics. From the reconstructed trajectories, potential MCSs (PMCSs) that subsequently develop into MCSs are mainly concentrated east of 85°E and south of 35°N in TP. Overall, 61.14% of TP MCSs propagate toward the east (eastward-leaning directions); relative to all TP MCS events, 1.85% move out of the plateau and 0.93% further enter Sichuan. A pronounced meridional gradient in TP MCS translation speed is indicated, with a mean speed of 60.50 km h⁻¹ north of 30°N, approximately 1.25 times that to the south. Based on ERA5 winds, statistically significant links between 200–400 hPa flow and MCS motion are identified: wind speed is weakly but significantly correlated with MCS speed (r = 0.18–0.20, 99% significance), whereas wind direction exhibits a stronger correlation with MCS movement direction (r = 0.29–0.34, 99% significance). A representative high-impact event was selected from TP-origin MCSs entering Sichuan. The case study indicates that these systems experienced significant enhancement upon entering Sichuan, attributable to merger with locally generated or stalled convection and to an environment with enhanced low-level moisture convergence, strengthened secondary circulation, and increased instability; these conditions intensify the coupling among ascent, condensation-related moisture sink, and latent heating, thereby improving organization and expanding the rainfall-affected area.