Scientific Reports最新文献

Technologies for recognizing facial attributes such as race, gender, age, and emotion from images of human faces have several applications, including personalized advertising, sentiment analysis, and the study of demographic trends and social behaviors. Analyzing face images and facial expressions presents several challenges due to the complexity of human facial attributes and the diversity in representation. While numerous attempts have been made to improve facial attribute classification performance, there remains a strong demand for enhanced accuracy. In this paper, we propose "FaceScanPaliGemma," a multi-agent vision language model (VLM) system consisting of four fine-tuned Google PaliGemma models, each specialized for a specific facial attribute classification. To evaluate the proposed solution, we used the public "FairFace" and "AffectNet" datasets. The results show high accuracy, reaching up to 81.1%, 95.8%, 80.0%, and 59.4% for race, gender, age group, and emotion classification, respectively, outperforming other VLMs such as OpenAI GPT, Google Gemini, LLaVA, and Google PaliGemma under zero-shot evaluation.

Calcium-activated chloride channels (CaCCs) are essential for epithelial secretion, neuronal transmission, and smooth muscle function. Among the Anoctamin family, Anoctamin 1 (ANO1) and Anoctamin 2 (ANO2) are classical CaCCs proteins. ANO1 has been identified as a potential therapeutic target due to its involvement in diseases such as cancer and cystic fibrosis. However, current high-throughput screening (HTS) systems face limitations in achieving subtype-specific detection and optimizing screening strategies. Stable HTS cell models expressing ANO1 or ANO2 were constructed via lentiviral transduction in Fischer mouse thyroid (FRT) cells. The models were validated using flow cytometry, Reverse Transcription Polymerase Chain Reaction(RT-PCR), and YFP-H148Q/I152L-based iodide fluorescence quenching assays. Patch-clamp electrophysiology was employed to characterize ANO1 and ANO2 current properties. Although these electrophysiological features have been previously reported, their application in HTS workflows had not been systematically evaluated. ANO1 displayed notable current rundown under sustained stimulation with high Ca²⁺ or agonist concentrations, whereas ANO2 maintained stable currents under identical conditions. Based on these findings, an optimized screening strategy was developed, incorporating agonist concentration gradients and the timing of inhibitor application. This approach improved the specificity and reliability of modulator detection. A robust and functionally validated cell-based HTS platform for CaCCs modulator discovery was established. By integrating the electrophysiological characteristics of ANO1 into the screening design, the optimized strategy enhances the accuracy of identifying selective ANO1 modulators. This work provides a methodological basis for future mechanism-driven screening of CaCCs-targeted compounds.

Early diagnosis of tuberculosis (TB) in children is hampered by non-specific symptoms, low bacillary loads, and difficulties in obtaining respiratory samples. We evaluated urinary metabolomics using high-resolution nuclear magnetic resonance (HR-¹H-NMR) spectroscopy as a non-invasive tool to discriminate TB disease, TB infection (TBI), and healthy controls (HC). Urine samples from 101 children enrolled in the Spanish Paediatric TB Network (pTBred) were retrospectively analysed, including 62 TB cases (26 microbiologically confirmed [cTB]), 17 TBI, and 22 HC. Metabolic fingerprints were generated via¹ H-NMR spectroscopy. Group discrimination was assessed using partial least squares discriminant analysis (PLS-DA) with cross-validation. Metabolites were selected using VIP > 2 and p < 0.05. Validated PLS-DA models (TB vs. HC, cTB vs. HC and cTB vs. TBI) achieved ≥ 70% accuracy (AUC-ROC 0.867-0.971), with sensitivities ranging from 69.1% to 90.0% and specificities from 77.3% to 86.7%. Discrimination was strongest for cTB vs. TBI (AUC-ROC 0.971; sensitivity 90.0%; specificity 86.7%). A set of discriminatory metabolites was identified across validated comparisons. Compared with HC, TB disease was characterized by increased levels of phenylalanine and two unidentified NMR signals, together with reductions in several energy and nitrogen-related metabolites; benzoate and histidine showed borderline differences. In contrast, TBI was associated with higher levels of isoleucine, N-acetylglutamine, glutamine, creatinine and 2-hydroxyvalerate. Urine HR-¹H-NMR metabolomics suggests potential distinct metabolic signatures across TB states. While not yet suitable as a confirmatory test, these preliminary findings show potential as a triage tool to help prioritise children for microbiological testing and may eventually reduce the need for invasive sampling.

This study delves into the antifungal capabilities and catalytic dye degradation efficiency of silver nanoparticles (AgNPs) synthesised using Enicostemma axillare leaf extract. Highlighting the pressing need for sustainable agriculture and pollution mitigation strategies, our investigation demonstrates the dual utility of phyto-synthesised AgNPs. Antifungal assays revealed substantial inhibition of mycelial growth and spore germination in Alternaria alternata and Fusarium oxysporum, with inhibition rates peaking at 83% and 82.79%, respectively, at a concentration of 120 µg/ml. Additionally, AgNPs showcased remarkable catalytic degradation of hazardous dyes methylene blue, methyl orange, and Congo red, achieving over 90% degradation within minutes, underlying their potential in environmental remediation. The catalytic reduction showcased not only the high efficiency of these nanoparticles in breaking down complex organic molecules but also positioned them as viable candidates for treating industrial effluents. The increased phenolic and proline contents in treated tomato plants suggest an enhanced stress response, potentially contributing to disease resistance. These findings underscore the versatility of phyto-synthesised AgNPs, offering promising avenues for the development of eco-friendly solutions in agriculture and environmental management.

Time-fractional diffusion equations have emerged as powerful models for describing anomalous transport phenomena in physics, biology and engineering. To address the computational challenges arising from their non-local operators, we employ the WEB-spline finite element method, which provides a flexible and accurate discretization framework. The resulting linear system of equations are then explored in the context of quantum computing. Specifically, we investigate three prominent quantum linear solvers: the variational quantum linear solver (VQLS), the Harrow-Hassidim-Lloyd (HHL) algorithm and quantum annealing (QA). VQLS leverages hybrid variational techniques and shallow circuits, making it well-suited for noisy intermediate-scale quantum (NISQ) devices, while HHL offers a theoretically exponential speedup for sparse systems but requires deep fault-tolerant circuits. QA, in contrast, reformulates the problem into a quadratic unconstrained binary optimization (QUBO) instance, enabling approximate solutions through energy minimization on specialized hardware. We present a comparative analysis in terms of circuit depth, noise resilience, scalability and solution extraction, including the role of quantum state tomography in reconstructing classical information. Numerical experiments on a time-fractional diffusion problem highlight the complementary strengths and limitations of each method. This study bridges advanced numerical discretization with emerging quantum algorithms, providing insights into the feasibility and future potential of quantum-enhanced solvers for fractional partial differential equations.