Search Results (100)

MASH is a progressive liver disease closely associated with cellular senescence, which is present in more than 80% of hepatocytes in patients who develop hepatocellular carcinoma (HCC). Although MASH affects both sexes, the incidence of MASH-related HCC is two to four times higher in males. Our group has previously described two apoptotic switches during MASH progression and HCC development, implicating the ATP1A1 signalosome in the late switch. Here, we investigated the role of ATP1A1 and sex-specific differences in the early apoptotic switch during preclinical MASH progression. Male and female C57BL/6J mice (7 weeks old) were fed normal mouse chow (NMC) or a high-fat diet (HFD) for 12, 24, or 48 weeks (n = 5/sex/group). Total body weight (TBW) and body composition were assessed by serial measurement and echo-MRI. Plasma was analyzed by non-targeted metabolomics and glutathione profiling using LC-MS/MS. NAFLD activity scores (NAS), hepatic senescence, and apoptosis were quantified in liver tissue. Statistical analyses were performed using GraphPad Prism and R. Males gained greater TBW and lean and fat mass than females (p < 0.05). At 24 W, males demonstrated higher GSH:GSSG ratios and lower ophthalmate levels than females (p < 0.05), consistent with altered redox balance. HFD-fed females showed increased succinic and deoxycholic acid levels, whereas males exhibited higher butyric acid levels across all time points (p < 0.05). Males had a higher mTOR 1 expression at 24 W and P53 at 12 W compared to females on HFD, but a lower Grb2 expression at 24 W (p < 0.05). By 24 W, males had lower fibrosis scores and reduced apoptotic activity compared with females (p < 0.05), despite similar levels of cellular senescence. The expression of ATP1A1, survivin, and SMAC did not differ by sex or diet, although an upregulation trend in both ATP1A1 and survivin was noted in the male-HFD group. There is sexual dimorphism in the response to HFD during the transition from senescence to the apoptosis-first apoptotic switch in MASH progression. Full article

Value factorization methods have become a standard tool for cooperative multi-agent reinforcement learning (MARL) in the centralized-training, decentralized-execution (CTDE) setting. QMIX (a monotonic mixing network for value factorization), in particular, constrains the joint action–value function to be a monotonic mixing of per-agent utilities, which guarantees consistency with individual greedy policies but can severely limit expressiveness on tasks with non-monotonic agent interactions. This work revisits this design choice and proposes Relaxed Monotonic QMIX (R-QMIX), a simple regularized variant of QMIX that encourages but does not strictly enforce the monotonicity constraint. R-QMIX removes the sign constraints on the mixing network weights and introduces a differentiable penalty on negative partial derivatives of the joint value with respect to each agent’s utility. This preserves the computational benefits of value factorization while allowing the joint value to deviate from strict monotonicity when beneficial. R-QMIX is implemented in a standard PyMARL (an open-source MARL codebase) and evaluated on the StarCraft Multi-Agent Challenge (SMAC). On a simple map (3m), R-QMIX matches the asymptotic performance of QMIX while learning substantially faster. On more challenging maps (MMM2, 6h vs. 8z, and 27m vs. 30m), R-QMIX significantly improves both sample efficiency and final win rate (WR), for example increasing the final-quarter mean win rate from $42.3\%$ to $97.1\%$ on MMM2, from $0.0\%$ to $57.5\%$ on 6h vs. 8z, and from $58.0\%$ to $96.6\%$ on 27m vs. 30m. These results suggest that soft monotonicity regularization is a practical way to bridge the gap between strictly monotonic value factorization and fully unconstrained joint value functions. A further comparison against QTRAN (Q-value transformation), a more expressive value factorization method, shows that R-QMIX achieves higher and more reliably convergent win rates on the challenging SMAC maps considered. Full article

This paper presents the Multi-agent Transfer Learning Based on Contrastive Role Relationship Representation (MCRR), focusing on the unique function of role mechanisms in cross-task knowledge transfer. The framework employs contrastive learning-driven role representation modeling to capture the differences and commonalities of agent behavior patterns among multiple tasks. We generate generalizable role representations and embed them into transfer policy networks, enabling agents to efficiently share role assignment knowledge during source task training and achieve policy transfer through precise role adaptation in unseen tasks. Unlike traditional methods relying on the generalization ability of neural networks, MCRR breaks through the coordination bottleneck in multi-agent systems for dynamic team collaboration by explicitly modeling role dynamics among tasks and constructing a cross-task role contrast model. In the SMAC benchmark task series, including mixed formations and quantity variations, MCRR significantly improves win rates in both source and unseen tasks. By outperforming mainstream baselines like MATTAR and UPDeT, MCRR validates the effectiveness of roles as a bridge for knowledge transfer. Full article

This study presents the first quantitative meta-analysis in cooperative multi-agent reinforcement learning (MARL). Undertaken on the StarCraft Multi-Agent Challenge (SMAC) benchmark, we quantify reproducibility and statistical heterogeneity across studies using the five algorithms introduced in the original SMAC paper (IQL, VDN, QMIX, COMA, QTRAN) on five widely used maps at a fixed 2M-step budget. The analysis pools win rates via multilevel mixed-effects meta-regression with cluster-robust variance and reports Algorithm × Map cell-specific heterogeneity and 95% prediction intervals. Results show that heterogeneity is pervasive: 17/25 cells exhibit high heterogeneity (I² ≥ 80%), indicating between-study variance dominates sampling error. Moderator analyses find publication year significantly explains part of residual variance, consistent with secular drift in tooling and defaults. Prediction intervals are broad across most cells, implying a new study can legitimately exhibit substantially lower or higher performance than pooled means. The study underscores the need for standardized reporting (SC2 versioning, evaluation episode counts, hyperparameters), preregistered map panels, open code/configurations, and machine-readable curves to enable robust, heterogeneity-aware synthesis and more reproducible SMAC benchmarking. Full article

Cooperative Multi-Agent Reinforcement Learning (MARL) relies on effective coordination among agents to maximize team performance in complex environments. However, existing coordination graph-based approaches often overlook dynamic group structures and struggle to accurately capture fine-grained inter-agent dependencies. In this paper, we introduce a novel method called the Group Attention Aware Coordination Graph (G2ACG), which builds upon the group modeling capabilities of the Group-Aware Coordination Graph (GACG). G2ACG incorporates a dynamic attention mechanism to dynamically compute edge weights in the coordination graph, enabling a more flexible and fine-grained representation of agent interactions. These learned edge weights guide a Graph Attention Network (GAT) to perform message passing and representation learning, and the resulting features are integrated into a global policy via QMIX for cooperative decision-making. Experimental results on the StarCraft II Multi-Agent Challenge (SMAC) benchmark show that G2ACG consistently outperforms strong baselines, including QMIX, DICG, and GACG, across various scenarios with diverse agent types and population sizes. Ablation studies further confirm the effectiveness of the proposed attention mechanism, demonstrating that both the number of attention heads and the number of GAT layers significantly affect performance, with a two-layer GAT and multi-head attention configuration yielding the best results. Full article

Coccidiosis due to Eimeria tenella remains a major constraint on the poultry industry. Previous studies have revealed that E. tenella infection triggers apoptosis in host cells. The mitochondrial permeability transition pore (MPTP) plays a pivotal role in the apoptosis and necrosis observed in infected host cells. However, the effect of MPTP opening on mitochondrial apoptotic factors remains unclear. To elucidate the dynamic changes in apoptotic signals during MPTP-mediated apoptosis in host cells infected with E. tenella, we established a chicken embryo caecal epithelial cell infection model. Cyclosporin A (CsA) was used to inhibit the MPTP. The infection rate was assessed by Hematoxylin and eosin (H&E) staining, whereas MPTP opening and the abundances of the mitochondrial apoptotic factors Smac, Endo G, and AIF were determined by flow cytometry and ELISA, respectively. The results revealed that both the degree of MPTP opening was markedly reduced in the E. tenella+CsA group compared to the E. tenella group (p < 0.05). Between 24 and 120 h post-infection (hpi), the cytoplasmic levels of Smac, Endo G, and AIF were significantly elevated in the E. tenella group compared with the control group (p < 0.05), while their mitochondrial levels were markedly decreased (p < 0.05). In contrast, mitochondrial expression of these factors was restored in the E. tenella+CsA group (p < 0.05), accompanied by a reduction in their cytoplasmic abundance (p < 0.05). These findings indicate that E. tenella promotes MPTP-dependent release of mitochondrial pro-apoptotic factors into the cytosol during the mid-to-late stages of infection, whereas pharmacological inhibition of the MPTP limits this redistribution. Full article

The Centralized Teacher with Decentralized Student (CTDS) framework is a multi-agent reinforcement learning (MARL) approach that utilizes knowledge distillation within the Centralized Training with Decentralized Execution (CTDE) paradigm. In this framework, a teacher module learns optimal Q-values using global observations and distills this knowledge to a student module that operates with only local information. However, CTDS has limitations including inefficient knowledge distillation processes and performance gaps between teacher and student modules. This paper proposes the evolutionary sampling method that employs genetic algorithms to optimize selective knowledge distillation in CTDS frameworks. Our approach utilizes a selective sampling strategy that focuses on samples with large Q-value differences between teacher and student models. The genetic algorithm optimizes adaptive sampling ratios through evolutionary processes, where the chromosome represent sampling ratio sequences. This evolutionary optimization discovers optimal adaptive sampling sequences that minimize teacher–student performance gaps. Experimental validation in the $StarCraft$ Multi-Agent Challenge (SMAC) environment confirms that our method achieved superior performance compared to the existing CTDS methods. This approach addresses the inefficiency in knowledge distillation and performance gap issues while improving overall performance through the genetic algorithm. Full article

Multi-agent reinforcement learning has demonstrated excellent performance in complex decision-making tasks such as electronic games, power grid management, and autonomous driving. However, its vulnerability to adversarial attacks may impede its widespread application. Currently, research on adversarial attacks in reinforcement learning primarily focuses on single-agent scenarios, while studies in multi-agent settings are relatively limited, especially regarding how to achieve optimized attacks with fewer steps. This paper aims to bridge the gap by proposing a heuristic exploration-based attack method named the Search for Key steps and Key agents Attack (SKKA). Unlike previous studies that train a reinforcement learning model to explore attack strategies, our approach relies on a constructed predictive model and a T-value function to search for the optimal attack strategy. The predictive model predicts the environment and agent states after executing the current attack for a certain period, based on simulated environment feedback. The T-value function is then used to evaluate the effectiveness of the current attack. We select the strategy with the highest attack effectiveness from all possible attacks and execute it in the real environment. Experimental results demonstrate that our attack method ensures maximum attack effectiveness while greatly reducing the number of attack steps, thereby improving attack efficiency. In the StarCraft Multi-Agent Challenge (SMAC) scenario, by attacking 5–15% of the time steps, we can reduce the win rate from 99% to nearly 0%. By attacking approximately 20% of the agents and 24% of the time steps, we can reduce the win rate to around 3%. Full article

Phosphate pollution in aquatic environments leads to eutrophication and harmful algal blooms, significantly impacting ecosystems and water quality. The current study evaluates the effectiveness of surface-modified activated carbon (SMAC) in suppressing phosphate release from sediments. Using soil samples from Daecheong Lake, the performance of SMAC adsorption for phosphate was analyzed under various SMAC modification scenarios. Experiments showed that SMAC achieved approximately twice the phosphate removal efficiency compared to conventional activated carbon, with increasing effectiveness under higher flow velocities. Additionally, SMAC significantly reduced phosphate concentrations within the sediment layers, proving its effectiveness in the soil remediation process as well. The results highlight SMAC as a promising solution for mitigating pollutant release in rivers, lakes, and coastal areas, offering both short-term and cumulative long-term benefits for water quality improvement and ecosystem protection. Full article

Construction is considered an ecosystem that incorporates many technical, legal, and practical aspects. This article focuses on the practical side of construction, specifically on the use of BIM (building information modeling) during the design phase. In the construction industry, BIM ensures better collaboration and project management by centralizing all information on the same platform. It allows each project stakeholder to exchange documents in real time and contribute to resolving any unforeseen issues or constraints that are encountered. Following an extensive literature review, it has been demonstrated that a lack of communication and collaboration among stakeholders is considered one of the major obstacles to the widespread use of BIM in architectural offices and agencies. The main contribution of this work is to clarify the roles of stakeholders and the rights and temporalities of access, consultation, and modification of the digital BIM model. To achieve this, a collaborative engineering method was applied to propose organizational models for the work of construction stakeholders who are using a collaborative engineering approach during the initial stage of a project (design phase), in order to define the roles of stakeholders and their access to the BIM model. Full article

HIV-1 infection is successfully treated by antiretroviral therapy; however, it is not curative as HIV-1 remains present in the viral reservoir. A strategy to eliminate the viral reservoir relies on the reactivation of the latent provirus to subsequently trigger immune-mediated clearance. Here, we investigated whether the activation of Toll-like receptor 8 (TLR8) or RIG-I-like receptor (RLR) together with the latency reversal agent (LRA) second mitochondrial-derived activator of caspases mimetics (SMACm) leads to HIV-1 reservoir reduction and antiviral immune activation. The TLR8 and RLR agonist elicited a robust pro-inflammatory cytokine response in PBMCs from both PWH and uninfected people. Notably, co-stimulation with SMACm specifically enhanced TLR8 induced pro-inflammatory cytokine as well as CD8 T cell responses. Ex vivo treatment of PBMCs from PWH with SMACm significantly decreased the size of the inducible HIV-1 reservoir, whereas targeting TLR8 or RLR reduced the HIV-1 reservoir in 50% of PWH ex vivo. Although co-stimulation with TLR8/RLR agonists further reduced the HIV-1 reservoir in 25% of PWH ex vivo, effectively inducing antiviral immunity may help eliminate reactivated HIV-1 cells in vivo. Our findings strongly suggest that LRAs can be used in combination with agonists for pattern recognition receptors to reactivate HIV-1 and induce antiviral immunity. Full article

Traditional multi-agent reinforcement learning (MARL) algorithms typically implement global parameter sharing across various types of heterogeneous agents without meticulously differentiating between different action semantics. This approach results in the action semantic conflict problem, which decreases the generalization ability of policy networks across heterogeneous types of agents and decreases the cooperation among agents in intricate scenarios. Conversely, completely independent agent parameters significantly escalate computational costs and training complexity. To address these challenges, we introduce an adaptive MARL algorithm named Generalized Action-Prediction Optimization (GAPO). First, we introduce the Generalized Action Space (GAS), which represents the union of all agent actions with distinct semantics. All agents first compute their unified representation in the GAS, and then generate their heterogeneous action policies with different available action masks. Second, in order to further improve cooperation between heterogeneous groups, we propose a Cross-Group Prediction (CGP) loss, which adaptively predicts the action policies of other groups by leveraging trajectory information. We integrate the GAPO into both value-based and policy-based MARL algorithms, giving rise to two practical algorithms: G-QMIX and G-MAPPO. Experimental results obtained within the SMAC, MPE, MAMuJoCo, and RPE environments demonstrate the superiority of G-QMIX and G-MAPPO over several state-of-the-art MARL methods, validating the effectiveness of our proposed adaptive generalized MARL approach. Full article

Introduction: Shiga toxin-producing Escherichia coli (STEC), particularly E. coli O157:H7, is a major contributor to foodborne outbreaks globally. Both E. coli O157 and non-O157 strains can lead to severe health issues, including hemolytic colitis and hemolytic uremic syndrome, which can result in kidney failure. Methods: Two hundred and one frozen beef samples were purchased from various supermarkets located in the Eastern Province of Saudi Arabia and subsequently enriched in tryptic soy broth (TSB). From the enriched samples in TSB, 1 mL portion was mixed with immunomagnetic beads (IMB) coated with specific antibodies targeting the E. coli O157 O antigen. The beads, which contained the captured bacteria, were then streaked onto CHROMagar O157 and Sorbitol MacConkey (SMAC) agar. The DNA extracted from these samples was examined using multiplex PCR to identify potential virulence gene markers, specifically stx-1, stx-2, and eae. Results: Of the 201 examined samples, 88 (43.8%) and 106 (52.7%) were positive for E. coli and produced colorless and mauve colonies on SMAC agar and CHROMagar O157, respectively. Out of 298 isolates in total, 174 isolates of E. coli were isolated with IMB enrichment. The highest detection rate of virulence gene markers was found among isolates that had been isolated using IMB enrichment, where 25 (8.4%), 2 (0.7%) and 12 (4%) isolates tested positive for stx1, stx2, and eae genes respectively. Among 42 isolates harboring potential virulence gene markers, 11 isolates were identified as E. coli O157 (stx₁⁺/eae⁺ or stx₂⁺/eae⁺). ERIC-PCR genotyping was able to determine the genetic relatedness among 42 isolates of E. coli O157 and E. coli non-O157 into 10 types with four identical related clusters and a genetic similarity rate above 90% homology from the identified isolates. Conclusions: The present study gives a clear perspective on STEC contamination in imported frozen beef marketed in Saudi Arabia. Because of the many possibilities of STEC contamination in imported frozen beef, further studies on the spread of STEC at various levels of imported frozen meat are needed on a long-term basis. Full article

Receptors for advanced glycation end products (RAGE) are multiligand cell surface receptors found most abundantly in lung tissue. This study sought to evaluate the role of RAGE in lung development by using a transgenic (TG) mouse model that spatially and temporally controlled RAGE overexpression. Histological imaging revealed that RAGE upregulation from embryonic day (E) 15.5 to E18.5 led to a thickened alveolar parenchyma and reduced alveolar surface area, while RAGE overexpression from E0 to E18.5 caused a significant loss of tissue and decreased architecture. Mitochondrial dysfunction was a hallmark of RAGE-mediated disruption, with decreased levels of anti-apoptotic BCL-W and elevated pro-apoptotic BID, SMAC, and HTRA2, indicating compromised mitochondrial integrity and increased intrinsic apoptotic activity. Extrinsic apoptotic signaling was similarly dysregulated, as evidenced by the increased expression of TNFRSF21, Fas/FasL, and Trail R2 in E0-18.5 RAGE TG mice. Additionally, reductions in IGFBP-3 and IGFBP-4, coupled with elevated p53 and decreased p27 expression, highlighted disruptions in the cell survival and cycle regulatory pathways. Despite the compensatory upregulation of inhibitors of apoptosis proteins (cIAP-2, XIAP, and Survivin), tissue loss and structural damage persisted. These findings underscore RAGE’s role as a pivotal modulator of lung development. Specifically, the timing of RAGE upregulation significantly impacts lung development by influencing pathways that cause distinct histological phenotypes. This research may foreshadow how RAGE signaling plausibly contributes to developmental lung diseases. Full article

Background: Percutaneous electrolysis is an invasive physical therapy technique that is receiving attention. The objective of this article is to evaluate the biological and cellular effects of percutaneous electrolysis and its influence on tissue healing processes. Methods. The search strategy performed in PubMed, Cochrane Library, and Web of Sciences databases resulted in a total of 25 studies. Once inclusion and exclusion criteria were applied, seven studies were finally included in this systematic review. The biological effects of percutaneous electrolysis were evaluated and grouped into pro-inflammatory and anti-inflammatory effects, cell death, and extracellular matrix and tissue remodeling effects. Results. Percutaneous electrolysis generates a significant pro-inflammatory increase in the chronic tendon condition of IL1β-6-18-1α-1rn, NLRP3, and M1 polymorphonuclear cells and increased expression of COX2, TNFα, Cxcl10, and TGFβ1 during the first 7 days. This inflammation is regulated as of day 13. A significant increase in cell death markers, such as LDH, Yo-Pro, cytochrome C, and Smac/Diablo markers, was observed during the first 7 days. Finally, a significant increase in markers Mmp9, VEGF, VEGFR, PPAR-γ/tubulin, and COL-I was observed in the extracellular matrix and tissue remodeling, and a decrease in COL-III was observed during the first 7 days. In the acute inflammatory injury condition, an increase in anti-inflammatory markers, such as IL-10-13, CCL1, and IkB, and a significant decrease in pro-inflammatory cytokines, such as IL-6-1β, CCL3-4-5, CCR5-8, NFkB, and TNFα, were observed during the first 7 days. Finally, a significant increase in VEGF, VEGFR, and PPAR-γ/tubulin markers in the extracellular matrix and tissue remodeling was observed for this condition during the first 7 days. Conclusions. Percutaneous electrolysis generates a controlled local pro-inflammatory effect in chronic conditions and regulates inflammation in inflammatory injuries (during the first 7 days). Electrolysis has short-term effects (0–7 days post) of cell death and controlled extracellular matrix destruction. Additionally, it facilitates subsequent healing by improving extracellular matrix synthesis starting from 7 days after application. Full article