Abstract: Background/Objectives: Plain radiography is the first-line imaging modality for acute foot and ankle trauma; however, computed tomography (CT) may provide additional information in selected patients with persistent clinical suspicion, equivocal radiographic findings, or a need for detailed fracture characterization. We hypothesized that CT positivity would be higher among patients with suspicious radiographs than among those with normal radiographs and that CT would identify additional fractures in a clinically selected cohort. Methods: This retrospective single-center selected-cohort study included 1000 adult patients with isolated foot and/or ankle trauma who underwent both plain radiography and CT during the same clinical encounter. Plain radiographs were categorized as normal, suspicious for fracture, or definite fracture on the basis of original radiology reports and available imaging records. CT positivity was defined as the presence of an acute fracture on CT. Analyses focused primarily on CT positivity rates and conditional diagnostic yield within the selected CT cohort. Conditional apparent diagnostic performance was evaluated as a supplementary sensitivity analysis and was not intended to estimate population-level diagnostic accuracy. Results: Plain radiographs were categorized as normal in 628 patients, suspicious in 283, and definite fracture in 89. CT detected acute fractures in 386 patients (38.6%). CT positivity rates were 14.8% in the normal radiograph group, 72.1% in the suspicious radiograph group, and 100.0% in the definite fracture group. The proportion of patients without a definite fracture on plain radiography but with an acute fracture on CT was 29.7%. Fracture-pattern classification was unavailable in 85 of 386 CT-positive cases (22.0%); therefore, fracture-pattern analyses were considered exploratory. Conclusions: In this selected CT cohort, CT identified additional acute fractures in a subset of adult isolated foot and ankle trauma cases that could not be definitively classified as fractures on plain radiography. Because CT was not systematically performed in all trauma patients, the findings represent selected-cohort CT yield rather than true population-level diagnostic accuracy.

Abstract: Background/Objectives: Healthcare students face occupational risk of hepatitis B vi-rus (HBV) exposure during clinical training, often many years after receiving their primary vaccination series in infancy or childhood. Progressive waning of an-ti-hepatitis B surface (anti-HBs) antibody titres, heterogeneous vaccination histories in international student cohorts, high rates of underreported needlestick injuries, and variable occupational health protocols across institutions raise important questions about the adequacy of pre-clinical immunization surveillance in this population. This review aimed to synthesize current evidence on hepatitis B immunization among healthcare students, focusing on occupational exposure risk, long-term vac-cine-induced immunity, clinical management of low or absent anti-HBs titres includ-ing post-exposure prophylaxis, and the specific challenges posed by multicultural aca-demic settings. Methods: A narrative review was conducted through a systematic lit-erature search of PubMed/MEDLINE and Scopus, covering publications from January 2000 to December 2025, supplemented by key seminal references. Search terms in-cluded combinations of "hepatitis B", "HBV vaccination", "anti-HBs", "healthcare stu-dents", "occupational exposure", "seroprotection", "long-term immunity", "needlestick injury", and "post-exposure prophylaxis". A total of 49 references were selected for in-clusion. Results: Needlestick and sharps injuries occur at measurable rates during clinical training, with medical students accounting for approximately 30% of occupa-tional exposure events in some series, and with underreporting rates estimated at 19–80% across studies. Pooled seroprotection prevalence at clinical training entry is ap-proximately 73.8% (95% CI 69.1–78.0%), with substantially lower rates among stu-dents vaccinated in infancy — as low as 28% at 16–20 years post-vaccination in longi-tudinal data — compared with adolescence. The majority of students with non-protective titres retain immunological memory, with 90.9% (95% CI 87.7–93.3%) demonstrating an anamnestic response after a single booster dose. Post-exposure prophylaxis pathways depend critically on pre-existing immunological status, rein-forcing the operational value of pre-clinical serological screening. International stu-dent cohorts present additional complexity due to heterogeneous vaccination sched-ules, documentation gaps, and variable natural immunity profiles. Conclusions: Sys-tematic pre-clinical serological screening encompassing anti-HBs, HBsAg, and an-ti-HBc, combined with evidence-based stepwise immunization management and structured educational interventions on safe sharps handling and reporting culture, is essential to protect healthcare students from occupational HBV exposure. Documented seroprotection status determines post-exposure prophylaxis decisions and, in some regulatory frameworks such as the Italian occupational health system, directly informs fitness for clinical duty assessments. These findings support a proactive, integrated approach to hepatitis B immunization surveillance in healthcare education, aligned with the WHO 2030 viral hepatitis elimination targets.

Abstract: This research answers the knowledge gap regarding the explanation of the quantum jump of the electron. This scientific paper aims to complete Einstein’s research regarding general relativity and attempt to link general relativity to quantum laws.

Abstract: Road infrastructure plays a strategic role in supporting the mobility of people and the distribution of goods, so that maintaining an adequate level of service requires effective and continuous maintenance programs informed by reliable pavement evaluation. In Indonesia, the functional performance of pavements is commonly assessed through the International Roughness Index (IRI) and the Pavement Condition Index (PCI); however, the relationship between these two indicators may differ from one road section to another depending on distress patterns and traffic characteristics. This study analyzes the correlation between IRI and PCI on eleven national road sections on Ambon Island, Maluku Province. Secondary data from the 2021 road condition survey conducted by the Maluku National Road Implementation Agency were analyzed using simple linear regression in IBM SPSS Statistics 23. The results reveal a moderate negative relationship, with a correlation coefficient (R) of 0.423 and a coefficient of determination (R²) of 0.179. The fitted model, IRI = 8.836 − 0.047 PCI, indicates that pavements in better condition (higher PCI) tend to exhibit lower roughness. The regression was statistically significant (F = 73.616; p < 0.001). Nevertheless, PCI explains only 17.9% of the variation in IRI, confirming that roughness is also governed by factors beyond visible surface distress, such as pavement age, traffic loading, and construction quality. The findings support the combined use of both indices for a more comprehensive functional assessment of urban national roads.

Abstract: CONTEXT Crop yield forecasting is essential for agricultural planning, food security assessment, and climate change adaptation. In Poland and other regions of Central Europe, interannual variability in early-season meteorological conditions is a key driver of yield variability, yet its systematic use in regional-scale forecasting models remains limited. OBJECTIVE This study aims to quantify the influence of early-season meteorological anomalies on yields of nine key field crops in Poland and to assess the potential for generating yield forecasts several months before harvest using satellite and reanalysis data. METHODS Linear regression models were developed using anomalies of 2-metre temperature, precipitation, reference evapotranspiration and root-zone soil moisture derived from ERA5 reanalysis and EUMETSAT H-SAF and LSA-SAF satellite products. Models were trained separately for each crop and each of the 16 Polish voivodeships using data from 2004 to 2022. A rolling time-based cross-validation scheme was applied, complemented by an independent hold-out validation for 2019–2022. RESULTS AND CONCLUSIONS Positive temperature anomalies during January–April were consistently associated with higher yields across all winter cereals, while soil moisture deficits in the upper root zone (SM1 and SM2 layers) represented the primary limiting factor. Reference evapotranspiration anomalies in May–June provided a proxy signal for radiation and atmospheric demand conditions. Preliminary forecasts were feasible up to four months before harvest for winter cereals and up to three months for spring cereals. Independent validation for the Opolskie voivodeship yielded Pearson correlation coefficients of 0.6–0.8 for most crops, with maize showing lower agreement (r ≈ 0.45). SIGNIFICANCE The proposed anomaly-based framework provides interpretable, timely and operationally relevant yield forecasts for regional agricultural planning under increasing climate variability. Its transparency and reliance on freely available satellite and reanalysis products make it suitable for implementation in early-warning systems across Central Europe.

Abstract: Ancient wheat materials are increasingly used in cereal-based products because of their perceived nutritional advantages and distinctive protein composition. However, their protein quality and IgE-binding properties after processing remain insufficiently characterized. This study evaluated total protein content, essential amino acid composition, ELISA-detectable gliadin content, and IgE-binding patterns in flours and corresponding pasta products prepared from eight industry-sourced ancient and modern wheat materials, including einkorn, emmer, Kamut, spelt, round grain wheat, wheat 2Ab, common wheat, and durum wheat. Protein content was determined by the Kjeldahl method, amino acid composition by UHPLC, gliadin content by direct ELISA, and IgE-binding properties by slot blot analysis using sera from allergic individuals. The analysed wheat materials differed markedly in protein content, amino acid profile, gliadin detectability, and IgE-binding capacity. Spelt and Kamut showed the highest total protein contents, while selected ancient wheat materials, particularly emmer, Kamut, and einkorn, exhibited more favourable scores for some essential amino acids than common wheat and durum wheat. Nevertheless, none of the analysed materials fully met the FAO/WHO reference pattern for all evaluated essential amino acids, with lysine and histidine remaining the main limiting amino acids. Einkorn flour showed the highest ELISA-detectable gliadin content, consistent with its distinctive gluten protein composition. Processing into pasta reduced ELISA-detectable gliadin levels in all analysed materials, indicating processing-related changes in protein extractability and epitope accessibility. Despite reduced gliadin detectability after processing, IgE-reactive components remained detectable in both flour and pasta extracts. IgE-binding patterns were strongly dependent on wheat material, product form, and individual serum profile. Kamut pasta repeatedly showed one of the strongest IgE-binding responses across different sera, indicating that processing did not eliminate IgE-reactive components in this material. These findings demonstrate that selected ancient wheat materials may offer favourable protein-related nutritional characteristics, but these features should not be interpreted as evidence of reduced IgE-binding capacity. Integrated assessment of protein quality, gliadin detectability, and IgE-binding properties is therefore necessary for a more complete evaluation of ancient and modern wheat-based products.

Abstract: Adversarial machine learning is a subdomain of trustworthy artificial intelligence, where machine learning and security overlap. Researchers in this field investigate the vulnerabilities of machine learning models, and aim to build more secure and robust systems against adversarial threats. Studies have shown that adversarial attacks in classical machine learning models transfer to quantum machine learning models, while new attacks emerge in quantum environments. In this survey, we study the attack surfaces, attacks, and defenses associated with the traditional machine learning models, neural networks, deep learners, GenAI, and agentic AI systems in quantum environments. As we move from traditional models to agentic AI systems, the attack surfaces increase. Increased attack surfaces provide novel opportunities for the adversaries, while attacks in the earlier paradigms are transferred to the newer ones. Moreover, we investigate the simulation platforms and datasets in the existing research studies, and provide practical research directions and guidelines for the future research.

Abstract: Coffee is one of the most widely consumed beverages worldwide. Coffee and its bioactive compounds, including caffeine, chlorogenic acid, and caffeic acid, have received increasing attention due to their potential contribution to coffee-associated health benefits. Evidence from pediatric and adult populations supports a positive association between fructose intake from sugar-sweetened beverages and the increasing prevalence of obesity and other non-communicable diseases. In this context, coffee consumption may represent a potential protective dietary factor against high fructose intake-induced metabolic alterations, including obesity, type 2 diabetes, liver disease, cardiovascular disease, and alterations in gut microbiota composition. However, human evidence on the effects of coffee consumption on fructose-related metabolic alterations remains limited, making it difficult to determine whether the benefits observed in animal models translate to humans. Therefore, this review summarizes current evidence on the potential role of coffee consumption and coffee-derived bioactive compounds in modulating fructose-induced metabolic alterations and discusses the mechanisms involved.

Abstract: Soybean is the main oilseed crop cultivated worldwide and is important for human and animal nutrition. The duration of flowering, reproductive period, and maturity is determined by genetic factors and influenced by environmental variables such as temperature, photoperiod, precipitation, soil type, and crop management, affecting yield and regional adaptation. However, gaps remain regarding the interaction between relative maturity groups (RMG) and the environment. Therefore, this study aimed to characterize the behavior of soybean populations and the environmental effect on the phenological cycle of the crop. Two populations were developed from biparental crosses with a narrow range in RMG among the parental cultivars. The E2 population originated from BMX Potência RR (RMG 6.7) × BMX Energia RR (RMG 5.3), and E3 from BRS 245 RR (RMG 7.3) × BRS 278 RR (RMG 9.4). The populations were evaluated over four growing seasons in an augmented block design, and traits were analyzed using mixed models with environmental covariates. Higher temperatures and lower precipitation were associated with shorter phenological periods, whereas higher precipitation and lower temperatures were associated with longer periods. Earlier lines were more environmentally sensitive, whereas later lines showed greater phenological stability.

Abstract: This paper presents a real-time browser-based platform for thermal‑hydraulic characteri-zation of a compact laboratory heat exchanger, implemented within the PolyVR re-search‑grade virtual reality engine. Experimental measurements—including temperature, flow rate, pressure, and electrical quantities—are retrieved from a cloud database at 1 Hz and processed through a browser‑native computational framework that performs contin-uous thermophysical modeling, hydraulic analysis, and energy‑balance evaluation. The system computes heat‑transfer rate, overall heat‑transfer coefficient, Reynolds, Prandtl, Nusselt, and Grashof numbers, pump performance, heater efficiency, and cumulative thermal energy. PolyVR serves as the runtime environment for the digital twin, providing modular integration of browser‑based computation with immersive 3D visualization. The platform supports animated flow diagrams, valve‑state indicators, thermal‑field moni-toring, and interactive manipulation of system components. Its scalable architecture ena-bles deployment across desktops, head‑mounted displays, and CAVE environments, with remote access facilitated through ngrok tunneling. Experimental validation using steady‑state and dynamic datasets demonstrates that the PolyVR‑enabled browser com-putation reproduces laboratory‑grade thermal‑hydraulic analysis with high fidelity, ena-bling accessible remote experimentation, immersive visualization of invisible thermal processes, and VR‑based digital‑twin interaction.

Abstract: Variant-based pathogenicity predictors such as REVEL evaluate missense variants in isolation, discarding the gene-length and allele-frequency context needed to compare collections of genes. In this paper, we introduce a composite gene-level metric integrating Hardy-Weinberg heterozygosity, coding-sequence length, and REVEL scores, evaluated on 55 high-confidence autism genes against the 1000 Genomes reference. It identifies elevated pathogenic burden in 48 of 55 genes, removes gene-length and variant-count confounds, and outperforms REVEL-only scoring by a large margin. Bootstrap resampling and a label-permutation control confirm the enrichment is stable and not an artefact of the scoring construction. We present it as a complementary gene-level layer for case-control and gene-set comparisons, with a nonlinear successor outlined as future work.

Abstract: Neck pain is a prevalent condition that affects a significant proportion of the global population, particularly among university employees. This study analyzes the relationship between low-density lipoprotein (LDL) levels and neck disability (ND) in university administrative and teaching staff. A cross-sectional study was conducted on 236 administrative and teaching participants from a private university in Quito, Ecuador. Most of the participants held administrative positions (64.83%). LDL levels were categorized based on the Adult Treatment Panel III guidelines, and ND was assessed using the Neck Disability Index (NDI). The results showed that 66.53% of participants had LDL levels classified as higher than desirable (near optimal or above), while most of the individuals (57.20%) exhibited mild disability. Ordinal logistic regression analysis revealed a weak inverse trend (β = −0.00444). This suggests a non-significant trend where higher LDL levels slightly correspond to a lower probability of being in a higher ND category. The p-value of 0.3095 indicates that the association was not statistically significant. Furthermore, a quadratic regression model showed that ND had negligible explanatory power over LDL variability (R² = 0.02599, p = 0.2039). In conclusion, this study found no statistically significant relationship between serum LDL levels and neck disability in university employees.

Abstract: Nowadays, the complexity of electronic health records (EHRs) requires tools capable of efficiently and accurately extracting and interpreting clinically relevant information to support clinicians. This study explores the use of the Cheshire Cat AI framework, configured with Ollama and using LLaMA3 as a language model, with the main purpose of performing automatic analysis of synthetic EHRs from Kaggle. Through specific structured queries, the model was able to successfully reconstruct patients’ clinical histories and extracted useful data such as diagnoses, treatments, visits, comorbidities and demographic data. A validation process through repeated queries was then performed, which confirmed a high level of accuracy. To preserve data privacy, only synthetic datasets were used in this work. Beyond the simple retrieval of information by means of queries, the study highlights the great potential of language models in clinical decision support. Their ability to interpret large and heterogeneous datasets certainly offers new opportunities to improve diagnostic accuracy, simplify workflows and personalise treatments. Specifically, natural language queries by tools such as Cheshire Cat AI can be used for intelligent support systems that can, for instance, integrate multimodal and real-time data to provide medical recommendations. These results represent a first step towards the exploitation of large language models not only for EHR analysis, but also to assist in clinical decision-making processes in different medical fields and, above all, for the study of specific complex diseases such as rare diseases.

Abstract: Flooding is the most recurrent and economically devastating natural hazard in Nigeria, yet no standard or consistent nationwide assessment method exists. Moreover, spatially explicit flood susceptibility information remains scarce—a critical gap for a country of over 220 million people with severely limited hydrometric monitoring infrastructure. This study presents one of the first nationwide, multi-model flood susceptibility mapping efforts for Nigeria, integrating four complementary approaches: Frequency Ratio (FR), Logistic Regression (LR), Random Forest (RF), and Gradient Boosting (XGBoost). The framework incorporates a bivariate FR component and Height Above Nearest Drainage (HAND) as a conditioning factor. Six conditioning factors were initially evaluated—elevation, TWI, HAND, LULC, slope, and soil type—with elevation, TWI, HAND, and LULC retained for final model development. The flood inventory was derived from a HEC-RAS 100-year floodplain simulation driven by satellite-derived discharge records from the Dartmouth Flood Observatory (DFO), yielding 973,111 binary flood/non-flood observations used as training labels for the LR, RF, and XGBoost models and as a flood-pixel count reference for FR computation. The HEC-RAS floodplain was independently verified against documented DFO historical flood reports. A three-tier accuracy assessment was conducted. For statistical accuracy using a 20% test subset, XGBoost achieved the highest AUC (0.956) and Overall Accuracy (0.892). For spatial consistency against the HEC-RAS reference, LR, RF, and XGBoost achieved substantial agreement (Kappa = 0.662–0.700), while FR achieved moderate agreement (Kappa = 0.410). Validation against the 2022 Sentinel-1 SAR flood extent showed that all four models exceeded HEC-RAS flood detection accuracy. For operational flood risk management, XGBoost is recommended due to its strong predictive performance and ability to minimize missed flood-prone areas. The resulting maps provide actionable spatial intelligence for disaster risk management, land-use planning, and early warning systems across Nigeria and other data-sparse regions.

Abstract: Fundamental physics faces three profound difficulties: ontological incompatibility ($G,c,\hbar$ presuppose different ontological categories), the chasm between continuity and discreteness (classical continuum vs. quantum discreteness), and divergent mathematical languages (differential geometry, Hilbert spaces, renormalization groups). This paper proposes a first-principles axiomatic system for the Cosmic Continuum (15 axioms A0–A14). A0 asserts: All state changes in the universe arise from energy redistribution; there is no state change without energy change. This axiom replaces Newton's "inertia" as the most fundamental physical assumption—inertia is merely the special case $\Delta E = 0$. Built upon this first principle, the axiomatic system is structured through ontological axioms (A1–A3) as the cornerstone, structural axioms (A4–A7) as the skeleton (including the connection axiom A4b which provides a unified description of the four fundamental interactions), dynamical axioms (A8–A12) as the realization (including the field recursion axiom A9b which unifies multiple fields as different levels of a recursive structure), and meta-axioms (A13–A14) as verification. The framework clearly distinguishes mass, energy, and dark mass beings and their corresponding space, time, and dark space dimensions, unified via the New Equivalence Principle (A3). The component (A4) serves as the skeleton, unifying particles and gauge fields into an inseparable tensor product. The scale topos and the $U_{q}(\mathfrak{e}_{8})$ modular tensor category provide the mathematical realization, rigorously capturing the relative continuum (A6) and embedding the Standard Model gauge group. The core breakthrough of this paper is to elevate ``interaction'' from an externally added Yang-Mills term to an intrinsic logic of component fusion. Gauge bosons emerge from the fusion decomposition $P \otimes P^*$ of the Planck particle $P$ ($\dim_q(P)=2$), with coupling strengths uniquely determined by fusion rules and braiding, not free parameters. Simultaneously, from A9 (fibred category) and A10 (2-category lifting), the quantum Boltzmann equation is rigorously derived, whose collision kernel is uniquely determined by the braiding $R$-matrix moduli and fusion coefficients, with no free parameters, achieving a complete axiomatic derivation from microscopic fusion rules to macroscopic transport phenomena. From these axioms we derive: the mirror 2-morphism $M$ is equivalent to CPT; the singularity is a phase boundary from ordinary spacetime to dark space; the singularity flux is quantized as $dN/dt = \mathrm{sgn}(-t)/t_{P}$; dark space entropy $S_{\mathrm{dark}} = k_{B}\ln 2\cdot N$ resolves the black hole information paradox; the mirror cyclic universe predicts $w_{a} > 0$ (dark energy weakens over time), consistent with current DESI/Planck data at $1.3\sigma$; wavefunction collapse is interpreted as a natural projection in the fibred category, with coherent information entering dark space. The framework is fundamentally deterministic (causal): the Born rule probabilities arise from limited access to information stored in dark space, not from intrinsic randomness. The framework proposes five testable predictions: (1) dark energy evolution direction $w_a>0$ ($w_a=0.12\pm0.09$); (2) CMB Planck-scale oscillations $\alpha\approx0.032$; (3) shear viscosity-to-entropy density ratio $\eta/s \gtrsim \hbar/(4\pi k_B)$; (4) LISA-band gravitational wave background peak $f_{\text{peak}}\approx0.2\,\text{Hz}$; and (5) black hole shadow quantum correction $\gamma\approx0.3$. This framework is the first to rigorously derive all four laws of thermodynamics within a single axiomatic system; the second law of thermodynamics is reframed as an apparent emergent phenomenon — the underlying dynamics are time-reversible and deterministic, while the apparent irreversibility arises from the observer's limited access to information in dark space. This paper presents a self-consistent, testable conceptual framework of axioms for fundamental physics, achieving a unification of classical physics, general relativity, quantum mechanics, thermodynamics, non-equilibrium statistical physics, and cosmology, thereby offering a fundamental response to the call for the axiomatization of physics raised by Hilbert's Sixth Problem.

Abstract: Large language models (LLMs) are being adopted quickly across biomedicine. Their value in clinical practice depends on more than accuracy: it also depends on whether they can run within the latency, hardware, privacy, cost, and staffing limits of real settings. In this scoping review of efficiency-oriented biomedical LLM research, we organize the literature along two axes: a taxonomy of efficiency techniques (prompting and retrieval, parameter-efficient and data-efficient adaptation, model compression, efficient architectures and inference, and agentic workflows) and a map of biomedical application domains. Across the corpus, prompting and parameter-efficient fine-tuning delivered efficiency most often, and studies reported it as savings in memory, trainable parameters, compute time, and human-workflow time, while energy and carbon were almost never measured. The reported gains are large and concrete: low-rank adaptation combined with low-precision quantization often shrinks the memory needed for adaptation enough to train and deploy a model on a single consumer or edge device, usually at a small and measured cost in task quality. Yet the evidence still leans toward retrospective benchmarking, with external validation, prospective evaluation, and clinical deployment all rare. We map which techniques serve which clinical domains, show how to read an efficiency claim against its comparator and clinical context, and identify what the field still needs to measure to turn demonstrated resource savings into validated clinical value.

Abstract: The identification of natural yellow dyes in ancient textiles is complicated, regardless of whether destructive or non-destructive techniques are used. The main limitation is the need for sampling, followed by deterioration and interference by the materials used in consolidation and restoration processes. In addition, different yellow dye sources share the same main fluorophores or components such as luteolin, kaempferol, and quercetin-based chromophores. We propose a minimally invasive methodology to identify sweet-scented marigold (Tagetes lucida), zacatlaxcalli (Cuscuta tinctoria), and weld (Reseda luteola). This methodology was tested on yellow wool samples that were dyed in an artisan workshop in the last 3 to 10 years. Another two samples of yellow wool fibres were obtained from the textile collection of the Franz Mayer Museum in Mexico City. Confocal scanning laser microscopy (CSLM), micro-Raman spectroscopy, attenuated total reflectance Fourier transformed infrared spectroscopy (ATR-FTIR), and variable pressure environmental scanning electron microscopy (VP-ESEM) were used to analyse the samples. The CLSM results showed that dyes are absorbed into the matrix of the fibres. The wool and dyes presented different emission spectra, which can be associated with the main groups of autofluorescent compounds in plants. The FTIR-ATR results supported the proteinaceous origin of the fibres, and the chemical composition and molecular structure of the autofluorescence phytocompounds were identified by micro-Raman spectroscopy. The findings indicate that the proposed methodology is adequate for identifying natural yellow dyes in wool fibres and can be applied to cultural heritage textiles.

Abstract: Background/Objectives: Nanvuranlat, a selective inhibitor of LAT1, has demonstrated clinical activity in advanced biliary tract cancer (BTC). We performed post hoc analyses to identify clinical and biomarker-defined populations that may derive greater benefit from LAT1 inhibition. Methods: Data from Phase 1 and Phase 2 studies were analyzed. Clinical outcomes were evaluated according to BTC subtype, prior primary tumor resection status, LAT1 expression, and accumulated drug exposure. Overall survival (OS), progression-free survival (PFS), and tumor size changes were assessed using Kaplan–Meier and Cox proportional hazards analyses. Results: In the Phase 2 study, nanvuranlat improved PFS versus placebo in the overall population (HR 0.56, 95% CI 0.34–0.90). Among the exploratory subgroups, lower hazard ratios were observed in patients without prior primary tumor resection (PFS HR 0.43, 95% CI 0.22–0.85; OS HR 0.53, 95% CI 0.28–1.01). Patients with high LAT1 expression demonstrated lower hazard ratios for PFS and OS than the overall population. Among patients with IHC, EHC, or GBC and high LAT1 2622551206500expression, PFS and OS hazard ratios were 0.31 (95% CI 0.15–0.64) and 0.50 (95% CI 0.25–1.00), respectively. Clinical outcomes also differed according to accumulated treatment exposure. Conclusions: These exploratory analyses suggest that BTC subtype, prior primary tumor resection status, and LAT1 expression may identify candidate populations for prospective evaluation in future nanvuranlat studies. The findings support further evaluation of biomarker-informed patient selection strategies in advanced BTC.

Abstract: This study investigates the dynamic transmission of geopolitical risk across the Brent crude oil market, gold market, U.S. Dollar Index (DXY), and the Thai stock market using a Bayesian Time-Varying Coefficient Vector Autoregressive (Bayesian TVC-VAR) model. Monthly data covering the period from January 1990 to December 2025 are employed to capture the evolving effects of major geopolitical events, including the Gulf War, the Asian Financial Crisis, the September 11 terrorist attacks, the Global Financial Crisis, the COVID-19 pandemic, and the Russia–Ukraine conflict. The analysis integrates Time-Varying Impulse Response Functions (TVIRFs), Generalized Forecast Error Variance Decomposition (GFEVD), the Total Connectedness Index (TCI), directional connectedness measures (TO, FROM, NET, and NPDC), and network analysis to examine both the magnitude and direction of shock transmission.The empirical findings indicate that geopolitical risk generates substantial time-varying spillover effects across commodity, foreign exchange, and equity markets. The intensity and direction of connectedness vary considerably across different geopolitical regimes, with oil and the U.S. dollar emerging as dominant transmitters of shocks during periods of heightened uncertainty, whereas gold primarily serves as a safe-haven asset that absorbs market disturbances. The Thai stock market exhibits greater vulnerability to external shocks during global crises, reflecting its high degree of integration with international financial markets. The network analysis further reveals that the topology of financial connectedness changes significantly during major geopolitical events, highlighting shifts in dominant transmission channels over time. This study contributes to the literature by providing a comprehensive Bayesian time-varying connectedness framework that simultaneously evaluates geopolitical risk, commodity markets, foreign exchange, and an emerging stock market. The findings offer valuable implications for investors, portfolio managers, central banks, and policymakers seeking to improve portfolio diversification, risk management, and financial stability under geopolitical uncertainty.

Abstract: Plant specialized metabolites sit at the boundary between plant genetics, environmental response, and useful natural products. Their accumulation is rarely constitutive; instead, it changes with tissue type, developmental stage, stress exposure, hormone signaling, and cellular storage capacity. In this review, we revisit basic helix-loop-helix (bHLH) transcription factors as regulatory switch points in plant specialized metabolism, with particular attention to the jasmonate-JAZ-MYC module. In resting tissues, JAZ repressors dampen MYC/bHLH activity. After wounding, herbivory, pathogen challenge, or elicitation, jasmonoyl-isoleucine promotes COI1-dependent JAZ turnover, freeing MYC factors to bind E-box/G-box motifs, recruit co-regulators such as MED25, and activate biosynthetic genes or downstream transcription-factor cascades. This logic has been repeatedly adapted in different plant lineages to regulate terpenoids, alkaloids, phenylpropanoids, flavonoids, glucosinolates, phytoalexins, and related metabolites. Examples discussed include Arabidopsis sesquiterpenes and glucosinolates, Taxus taxanes, Artemisia artemisinin, Catharanthus terpenoid indole alkaloids, Salvia phenolic acids and tanshinones, Ginkgo terpene trilactones, rice diterpenoid phytoalexins, and cotton gossypol. Rather than treating bHLHs as stand-alone master regulators, we frame them as context-dependent nodes whose outputs depend on dimer choice, promoter grammar, chromatin accessibility, hormone crosstalk, partner transcription factors, and cell-type competence. We also outline evidence standards and engineering principles for using bHLH switches in crop defense, food-quality improvement, medicinal-plant production, and synthetic biology.