Search Results (2,257)

Computational fluid dynamics (CFD) workflows remain strongly dependent on expert decisions in geometry preparation, meshing, model and solver configuration, convergence assessment, post-processing, and design iteration. This review examines agentic artificial intelligence (AI) for CFD automation and engineering design, with emphasis on reinforcement learning (RL) agents and large language model (LLM) agents. RL agents provide rapid, task-specific, and physics-aware execution within predefined observation and action spaces, whereas LLM agents offer flexible goal interpretation, knowledge access, planning, and tool orchestration. To organize these potentially complementary capabilities, a unified taxonomy is introduced based on decision paradigm, CFD role, and degree of autonomy, while multi-agent organization is treated as a separate architectural property. The reviewed evidence indicates that neither RL nor LLM agents alone are sufficient for trustworthy autonomous CFD. A hierarchical autonomous-CFD-engineer architecture is therefore proposed, combining an LLM orchestrator, domain-specialized agents, conventional CFD tools, independent numerical and physical verification, shared provenance, and human governance. Extending this architecture toward hypothesis formulation, numerical-experiment design, and model discovery suggests a pathway to an AI CFD scientist; however, current systems support bounded automation and supervised scientific collaboration rather than independent scientific judgment. Full article

Retrieval-augmented generation and structured agent memory are commonly treated as competing ways to ground large language models. Our results show that their relative value changes with the task and that neither retrieved similarity nor graph association is sufficient evidence for an answer. We present a task-adaptive two-layer methodology combining a compact, reviewed knowledge layer for orientation and cross-source synthesis with a versioned evidence layer for exact lookup and claim support. A router selects the representation, while an independent admission stage checks named entities, time, source role, contradiction, and provenance. A controlled course-grounding study reveals a task-conditional crossover within Course 2: the distilled Body of Knowledge (BoK) trails plain RAG by 1.07 points on lookup and leads by 0.62 points on synthesis. Across a 72-question three-course lookup panel, RAG leads by 2.17 points; Course 1 synthesis is near parity. Complementary, position-debiased LLM-judged mastery comparisons favor the BoK on both courses. The standing representation is reduced 22–35-fold. An observational production case over a 289,642-chunk evidence base illustrates why candidate discovery, packet selection, and evidence admission require separate gates. The contribution is a tested method for assigning retrieval, structure, and evidence admission distinct responsibilities within one grounded agent architecture. Full article

The reciprocal relationship between large language models (LLMs) and knowledge graphs (KGs) opens new opportunities for knowledge discovery from scientific literature; however, most approaches address only one direction of this interaction. In complex biomedical domains, ensuring reliability, interpretability, and traceability remains challenging, making expert supervision essential. In this work, we propose a framework that integrates LLMs, KGs, and a Human-in-the-Loop scheme for constructing, evaluating, and reasoning over Evidence-based Knowledge Graphs (E-KGs), explicitly linking each extracted triplet to supporting textual evidence, and demonstrate its application to study the phenomenon of resistance in ovarian cancer. The E-KG was constructed using MedGemma-27B, with experts participating throughout the E-KG lifecycle, supported by our interactive visualization tools. We evaluated the E-KG through an expert-driven protocol assessing coherence and biological plausibility with inter-annotator agreement (Gwet’s AC1 coefficient), complemented by GraphRAG-based tasks, including link prediction (Hits@k, MRR) and a question-answering case study (faithfulness and answer relevance by LLM-as-a-judge and expert audit). The framework produced a coherent, plausible, and auditable E-KG, highlighting the potential of evidence-grounded knowledge to support LLM-based reasoning. To our knowledge, we introduce the first E-KG specialized in ovarian cancer resistance, providing an expert-validated resource that advances trustworthy and explainable AI for biomedical knowledge discovery. Full article

Late-onset Alzheimer’s disease (LOAD) is a cognitive disorder that leads to the requirement of full-time care. The ever-increasing growth of the LOAD population demands that more research be dedicated to the discovery/design of effective therapeutics. The two major hypotheses that have been evaluated in the clinic include the cholinergic and the amyloid hypotheses. Unfortunately, the therapeutics that have resulted from research in these areas have minimal clinical impact. These are, respectively, the cholinesterase inhibitors and the anti-amyloid therapies. Clearly, a new path forward is needed. To this end, a number of genetic approaches and biochemical investigations are currently being employed to provide new insights into the pathophysiology of dementia. While there currently is a diverse array of dysfunctions that have been reported in LOAD subjects, white matter (WM) dysfunction may be a nodal point that is critical to the pathophysiology of cognitive decline. In this review, I present our current knowledge of WM dysfunction in LOAD. This addresses the potential contributions of microvascular dysfunction to WM abnormalities and neuronal dysfunction. This knowledge base has the potential to lead to a new approach for treating cholinergic dysfunction in cognitive disorders. Full article

Science is dynamic, and novel models to better understand experimental observations are very important for its advancement. In the context of cell biology, knowledge about how cells behave in response to extracellular stimuli, and how these processes are dysregulated in pathologic conditions, is one of the challenges of biological research nowadays. From a theoretical point of view, the discovery of the double-helix model of DNA structure led to the idea that most aspects of a cell’s behavior could be understood by knowing the sequence and the rules regulating the expression of genes. A paradigm of this way of thinking is the theory of the “selfish gene” by Richard Dawkins. Indeed, this approach is being found to have several limitations, especially in the interpretations of complex pathologies, such as cancer, metabolic or neurodegenerative diseases. For this reason, a paradigm shift should be considered to provide a better modeling of the biological processes that control cellular behavior and physiology. In particular, a central role should be given to metabolism as a relay between the environment and microenvironment and cell function, by driving epigenetic modifications. For these reasons, a new “selfish cell” model should be introduced, which could prove useful to better explain complex diseases and find more effective translational approaches. Full article

Mexico is a global hotspot for plant diversity, yet its floristic inventory remains incomplete. Understanding whether taxonomic discovery is advancing evenly is essential for guiding future work. Here, we ask what the taxonomic, geographic, and methodological patterns of species proposals reveal about the current state of and persistent gaps in Mexican floristics. To address this, we synthesized the taxonomic, geographic, and methodological patterns of all new embryophyte species published for Mexico from 2000 to 2025. We compiled a database of 2822 new extant and fossil species, showed an average annual rate of 108.3 species, and analyzed collection records, authorship, institutional participation, publication timelines, and diagnostic evidence. The most diverse families were Asteraceae (278), Orchidaceae (230), and Fabaceae (202), while Tillandsia (69) and Salvia (67) were the most prolific genera. Oaxaca, Chiapas, and Jalisco concentrated the highest number of novelties. National collectors (566) and authors (520) played a predominant role, with Mexican institutions, led by Universidad Nacional Autónoma de México, Instituto de Ecología, and Universidad de Guadalajara, contributing the largest proportion of species; 93% of holotypes are housed in national herbaria, which contrasts with the historical pattern in other countries. Morphological data alone supported 80% of proposals, and only 2.5% explicitly stated a species concept. While 61% of species were published within 20 years of collection, some took over a century to be published. Our findings reveal sustained progress and strong national leadership in floristic knowledge, but also highlight persistent geographic, taxonomic, and editorial asymmetries. We emphasize the urgent need to strengthen taxonomic infrastructure, digitize collections, ensure generational turnover, and support Mexican scientific journals to secure the country’s taxonomic sovereignty. Full article

Background and Objectives: Defective resolution of inflammation may contribute to chronicity in atopic dermatitis (AD). Maresin-1 (MaR1) and Maresin-2 (MaR2) are docosahexaenoic acid-derived specialised pro-resolving mediators, and adipolin/C1q/TNF-related protein 12 (CTRP12) is an adipokine with anti-inflammatory properties; to our knowledge, none has been evaluated in the circulation of patients with AD. Materials and Methods: In this single-centre case–control study, 44 adults with AD and 44 healthy controls were enrolled. Serum MaR1, MaR2, and adipolin were measured by enzyme-linked immunosorbent assay, and disease severity was graded with the Scoring Atopic Dermatitis (SCORAD) index and the Eczema Area and Severity Index (EASI). Between-group comparisons were complemented by covariate-adjusted models (age, sex, body mass index, smoking), Spearman correlation, receiver operating characteristic (ROC) analysis, and false discovery rate control. Results: Serum adipolin (3.16 vs. 3.23 ng/mL; p = 0.211), MaR1 (140.31 vs. 141.09 ng/L; p = 0.815), and MaR2 (67.56 vs. 72.29 pg/mL; p = 0.512) did not differ between patients and controls, and the null findings persisted after adjustment (all p ≥ 0.306). None of the three biomarkers correlated with SCORAD or EASI (all p ≥ 0.389) or discriminated patients from controls (area under the ROC curve 0.459–0.578). Results were unchanged when measurements outside the assay calibration ranges (adipolin, 67.0% below; MaR1, 14.8% and MaR2, 13.6% above) were censored. Serum lactate dehydrogenase was higher in patients (217.52 vs. 183.39 U/L; q = 0.004) and correlated with SCORAD (ρ = 0.404) and EASI (ρ = 0.349). Adipolin and MaR1 were strongly correlated (ρ = 0.815), including among samples with adipolin below the calibration range (ρ = 0.640). Conclusions: Circulating MaR1, MaR2, and adipolin/CTRP12 were not altered in adults with AD and do not appear useful as standalone diagnostic or severity biomarkers, although compartmentalised cutaneous activity cannot be excluded. The adipolin–MaR1 correlation showed features of an analytical artefact and should not be interpreted biologically. Full article

This review synthesizes current knowledge of Enceladus as an Ocean World, with emphasis on habitability, organic chemistry, and the search for life. We summarize data obtained by the Cassini spacecraft, including the discovery and characterization of Enceladus’ plume and subsurface ocean. We assess how plume materials constrain the composition of the ocean, while highlighting uncertainties introduced by fractionation through transport and possible ice–ocean interface processes. We further examine how ocean dynamics and plume–ocean connectivity may shape the transport and transformation of dissolved and particulate species from the moon’s interior to space. We consider possible sources of chemical energy, redox disequilibria, and abiotic organic carbon, and how they could constrain ocean biomass, biological productivity, and the detectability of potential biosignatures. Finally, we evaluate near-term exploration strategies and mission architectures for testing theories of habitability, chemical evolution, and possible life beyond Earth. Full article

Spondyloarthritis is a multi-faceted condition housing a group of inter-related chronic inflammatory rheumatic diseases such as ankylosing spondylitis, IBD-associated arthritis, psoriatic arthritis and juvenile idiopathic arthritis. While decades of research and technological advancements have significantly improved our understanding of the disease, there remain many underlying genetic and physiological mechanisms which are unsolved but essential for the development of the right diagnosis and treatment. This mini-review examines our current foundational knowledge of spondyloarthritis, the main concepts related to the disease pathogenesis and the various lines of therapies developed thus far. We hope to provide a perspective on the current and potential antibody discovery platforms for the next decade as computational methods and artificial intelligence become increasingly mature and relevant in clinical research. Full article

Identifying bioactive compounds in traditional Chinese medicine (TCM) requires computational prioritization to be connected with chemical and pharmacological evidence. This structured narrative and critical review examines machine learning applications in compound prediction, formula-compatibility analysis, multi-omics integration, and drug development and repurposing. Molecular fingerprints, graph and sequence representations, knowledge graphs, and generative models are discussed in relation to their intended endpoints and validation conditions. Studies are classified along two independent dimensions: application scope, distinguishing direct TCM applications from natural-product studies and general drug-discovery methods, and validation level, ranging from computational prediction to chemical, experimental, and clinical evidence. Three case analyses illustrate a TCM-specific graph-learning platform with chemical confirmation, a downstream animal and metabolomics study without machine learning screening, and a proposed causal framework assembled from separate studies. Across these applications, machine learning supports candidate ranking and hypothesis generation, but predicted targets, attention weights, and multi-omics associations do not independently establish target engagement, formula synergy, or therapeutic efficacy. Major limitations include chemical-domain shift, inconsistent database provenance, data leakage, sparse negative results, and incomplete experimental follow-up. Reliable advancement requires scaffold-aware or external evaluation, explicit applicability domains and uncertainty, traceable data processing, and constituent-resolved intervention studies. A minimum reporting checklist consolidates these requirements, while route-specific assessment distinguishes synthetic accessibility from biosynthetic feasibility. Together, these considerations provide a basis for evaluating screening claims and designing reproducible workflows toward experimentally supported TCM compound discovery. Full article

Background/Objectives: Minimally invasive endodontics (MIE) has gained increasing attention as an approach focused on preserving dental tissues while achieving the objectives of endodontic treatment. This study compared self-perceived knowledge, perceived adequacy of education, self-reported clinical use and practices, implementation difficulties, and barriers to routine use of MIE approaches among general dentists and endodontists in Türkiye. Methods: This online cross-sectional survey used non-probability voluntary sampling and included 300 participants (150 general dentists and 150 endodontists). Group comparisons used Pearson’s chi-square test, Yates-corrected Pearson’s chi-square test, or Fisher’s exact test as appropriate. For multiple-response items, option-specific comparisons were adjusted using the Benjamini–Hochberg false discovery rate (FDR) procedure. Associated factors were evaluated using binary logistic regression, with professional group, professional experience, institutional setting, and weekly number of root canal treatments included simultaneously in multivariable models. Results: Endodontists reported higher self-perceived knowledge of MIE and more frequent self-reported use of MIE approaches than general dentists (both p < 0.001). In multivariable models, endodontist status was associated with higher odds of reported use of vital pulp therapies (odds ratio (OR) = 5.626) and minimally invasive root canal preparation (OR = 2.238), whereas the association with minimally invasive access cavity use was not statistically significant (OR = 1.668; p = 0.200). General dentist status was associated with higher odds of reporting difficulties during vital pulp therapies (OR = 10.420) and minimally invasive root canal preparation (OR = 2.379), whereas no significant association was detected for minimally invasive access cavity difficulties. The use of magnification systems and specialized endodontic burs for access cavity preparation or refinement was reported more frequently among endodontists. Reduced loss of tooth structure was the most frequently selected reason for using MIE, while inadequate training or lack of knowledge was the most frequently reported barrier to routine implementation. Conclusions: Among the participating dentists, endodontists reported higher self-perceived MIE knowledge and more frequent self-reported clinical use than general dentists. Professional and practice-related characteristics were associated with several MIE-related outcomes. Greater emphasis on practical, case-based training may support the appropriate implementation of MIE; however, the findings should be interpreted in light of the cross-sectional, self-reported design. Full article

Background/Objectives: Chronic Obstructive Pulmonary Disease (COPD) is a progressive respiratory disorder imposing massive global mortality and socioeconomic burden, with disproportionately high prevalence and suboptimal diagnostic rates in rural China. This study aimed to identify factors associated with two-dimensional diagnostic adherence—operationalized as a cognitive/psychological dimension (knowledge of diagnostic guidelines, perceived value of early diagnosis, and psychological readiness) and a behavioral dimension (completion of standard diagnostic procedures)—among COPD high-risk populations based on the Health Belief Model (HBM), and to explore the statistical mediating roles of health beliefs and social support. Economic toxicity refers to the financial burden of healthcare, including direct costs (copayments, medications) and indirect costs (transport, lost wages). Methods: A cross-sectional survey was conducted among 248 COPD high-risk individuals recruited via stratified random sampling in Pingdu District, Shandong Province, eastern China. A two-stage screening procedure was employed: first-stage screening at township health centers using portable spirometry, followed by referral for standard diagnostic evaluation at secondary or higher-level hospitals. Validated scales were applied for data collection, including the European Health Literacy Survey Questionnaire-Short Form (HLS-EU-Q16). Analyses included descriptive statistics, reliability/validity testing, Pearson and point-biserial correlations, multiple linear regression (cognitive outcome), multivariable logistic regression (behavioral outcome), and Hayes’ PROCESS macro with 5000 bootstrap samples. Benjamini–Hochberg false discovery rate correction was applied to multiple comparisons. Results: The mean cognitive dimension score was 4.935 ± 2.512, and the behavioral diagnostic completion rate was 42.3%. Independent predictors of cognitive adherence included health literacy (β = 0.286), health beliefs (β = 0.252), COPD knowledge (β = 0.239), living alone (β = −0.157), and economic toxicity (β = −0.108). Predictors of behavioral adherence identified via logistic regression were health literacy (OR = 1.28, 95%CI: 1.15–1.42), social support (OR = 1.12, 95%CI: 1.04–1.21), economic toxicity (OR = 0.82, 95%CI: 0.72–0.94), and medical accessibility (OR = 1.85, 95%CI: 1.12–3.05). Health beliefs statistically mediated 31.4% of the knowledge → cognitive adherence association and 36.9% of the economic toxicity → cognitive adherence association (cognitive outcome; behavioral indirect associations are reported on the log-odds scale with bootstrap confidence intervals). GAD-7 anxiety score was weakly correlated with adherence but not an independent predictor. Conclusions: Health literacy, health beliefs, and COPD knowledge are associated with higher COPD diagnostic adherence in rural high-risk populations in eastern China, while economic toxicity and living alone are associated with lower adherence. Health beliefs mediate the statistical associations between knowledge and economic status and cognitive adherence. Given the cross-sectional design, these findings should be interpreted as associations rather than causal effects. Dual-target interventions covering both cognitive and behavioral barriers are proposed as future strategies to optimize early COPD diagnosis in rural primary care. Full article

The transition toward a circular economy requires firms to develop circular business models (CBMs), yet doing so is particularly challenging for ventures operating under severe resource constraints and uncertainty. Although prior research has examined how entrepreneurs mobilize, recombine, and redeploy resources, less is known about how successive expansions of uses of the same focal resource become generatively connected over time. In this study, I address this question through an in-depth case study of TOMUSHI Inc., a Japanese venture that has developed multiple businesses around a single biological resource—the rhinoceros beetle. I propose a Cyclical Model of Expanding Uses of a Focal Resource. The model captures two analytically distinct but interrelated processes: a within-cycle process through which a new use of the focal resource emerges and a between-cycle process through which the outcomes of one cycle create conditions for subsequent expansion. The findings show that the enactment and commercialization of one use generated knowledge, stakeholder relationships, legitimacy, and capabilities that subsequently became means or conditions for further expanding other uses of the same focal resource. The theoretical significance lies not simply in the identification of multiple uses but in the discovery of generative connections between successive uses. By unpacking this process, I provide a fine-grained account of means expansion within effectuation and show how successive episodes of means expansion can become generatively connected across cycles. Full article

The smart specialization strategy (S3) is a place-based approach to regional development, which rapidly became highly demanded across regions regardless of its novelty. In essence, S3 aims to support regions to transform their economic structures in favor of knowledge-based development. Practical implementation includes setting a limited number of priorities within development strategies and involving a wide range of stakeholders. The S3 approach differs from conventional innovation policies because it focuses on the process of “self-discovery” and targets a limited number of activities that have the potential to serve as a regional competitive advantage. Today, the approach continues to evolve. The updated vision of regional development accentuates sustainability at the core of regional growth. To achieve better outcomes, regions need to localize the Sustainable Development Goals (SDGs) and discover opportunities based on their own capacity, potential, and vision. Smart specialization strategy for sustainability (S4) aims to embed sustainability into economic activities, aligning them with local characteristics and challenges. The key objective of this transformation-oriented regional development policy is to help regions become places with high business activity and an attractive environment in which to work, live, and invest. A special contribution of S3 and S4 is their support in discovering regional potential, even in declining regions. This can significantly contribute to closing productivity and income gaps. Full article

Corals of the genus Porites Link, 1807 are often considered difficult to identify to species level, largely due to their complex and historically unstable taxonomy. Much of this uncertainty stems from a long history of inconsistent species descriptions and revisions, which have blurred diagnostic boundaries between taxa. This is compounded by their very small corallite size, which limits the resolution of traditional morphological approaches and has historically precluded detailed morphometric comparisons. Together, these factors have resulted in persistent taxonomic ambiguity within the genus, making species-level identification challenging using morphology alone. Here, we take a two-phase approach to synthesising the taxonomic history of ecologically important massive species within the genus by revisiting original descriptions and reimaging type material to provide more consistent interpretations of morphological features and distinguishing characters between species. We focus on ten massive-forming species from the Indo-Pacific and provide the most comprehensive review of the full nomenclatural history of these species, along with the wider genus itself. We find that at least 22 characters are diagnostically informative for species-level discrimination within the type material examined, although their diagnostic value varies among species and character combinations. Importantly, wall architecture is one of the most useful characters, broadly separating the species into two groups: those with perforate (blistered) walls and those with reticulate walls. However, operationalising Porites taxonomy to robustly resolve species boundaries and refine estimates of species richness beyond type localities will require well-replicated integrative studies that combine molecular evolutionary evidence with ecological and reproductive traits at both regional and transoceanic scales. By building on the legacy of earlier systematists, this work fills a key knowledge gap and provides a robust morphological foundation for advancing Porites taxonomy, supporting more accurate species discovery and identification into the future. Full article