Browse Articles
Discover research articles across all indexed journals
Quantum-inspired quadratic attention with fourier-domain rank control in transformer architectures
Electron tomography reveals mitochondrial network and cristae remodelling during cell differentiation in the human placenta
Abstract Mitochondria adapt their structure through fusion and fission, yet how their morphology and cristae architecture change as cells differentiate remains unclear. The human placenta is a valuable model for studying mitochondrial diversity within a single tissue. As epithelial trophoblast cells of the placenta differentiate, their accompanying mitochondria morphologically and functionally transform. We use array tomography to characterise mitochondrial volume and network complexity. Cryo-electron tomography reveals two distinct subpopulations of mitochondria within preparations enriched for progenitor cytotrophoblasts, and a singular, homogeneous population in the differentiated syncytiotrophoblast. We showcase the 3D topology of individual cristae through a standardised metric of “curvedness”. Proteomic analysis of isolated mitochondria identifies reduced levels of proteins involved in mitochondrial dynamics and cristae organisation complexes in the syncytiotrophoblast, accompanied by variations in electron transport chain subunits, ATP synthase, and supercomplex components. This study highlights the advantages of combining multimodal imaging with proteomic analyses, to elucidate the process of mitochondrial morphology and cristae architecture remodelling as cells differentiate.
Retraction: Sports resources utilization efficiency, productivity change, and regional production technology heterogeneity in Chinese Provinces: DEA-SBM and Malmquist Index Approaches
Korean lunar navigation constellation with stability-filtered optimisation
Abstract The international lunar navigation baseline—eight satellites in elliptical lunar frozen orbits (ELFOs) with $$\omega =90^{\circ }$$ —provides robust south-polar PNT coverage but leaves near-side mid-latitudes largely unserved. This gap directly affects the Korea Aerospace Administration (KASA) 2032 lander mission, whose candidate sites span latitudes $$40^{\circ }$$ – $$70^{\circ }$$ . We propose the Korean Lunar Navigation System (KLNS): five augmentation satellites in $$\omega =270^{\circ }$$ ELFOs that concentrate dwell time over the northern hemisphere. A two-stage optimisation is employed. First, every candidate orbit in a 10,368-element search space is propagated for 720 hours under a perturbed model (lunar $$J_2$$ + Earth third-body) and screened against four stability criteria, removing 2864 (27.6%) dynamically infeasible candidates. Second, a greedy sequential algorithm with hard S-ROI protection selects five KLNS satellites that raise N40–70 PNT availability from 0.0 to 89.5%, far exceeding the 50% target, while preserving 99.7% south-polar coverage. The stability-filtered greedy solution outperforms a $$100\times 50$$ genetic-algorithm benchmark (85.6%), demonstrating that a well-designed pre-filter can obviate more expensive metaheuristics. Site-specific evaluation confirms 92.5–94.2% availability at all three northern KASA candidate landing sites under the augmented constellation.
A metabolite-bridged complex between SerRS and SIRT2 couples NAD⁺ metabolism to translation control
Abstract Cellular homeostasis requires tight coordination between metabolic and translational networks. Here we identify a direct molecular link between these processes through a cryo-EM structure of human cytosolic seryl-tRNA synthetase (SerRS) in complex with the NAD⁺-dependent deacetylase SIRT2. This interaction is promoted by the NAD⁺ metabolite ADP-ribose (ADPR), which acts as a molecular bridge between the two enzymes. Within the SIRT2 active site, ADPR engages SerRS residue K414 located in a flexible catalytic-domain loop. Acetylation of K414 is dispensable for binding. Functionally, complex formation inhibits SIRT2 deacetylase activity by blocking substrate access, while SIRT2 association suppresses SerRS aminoacylation activity by preventing tRNA binding. Thus, SerRS and SIRT2 mutually regulate each other, with ADPR enhancing while tRNA attenuating their interaction. Oxidative stress promotes this interaction via a PARP1-dependent pathway, revealing an ADPR-responsive regulatory module that couples metabolic state to translational output. This regulatory module is likely conserved across vertebrates.
Predicting T790M mutation status in non-small cell lung cancer based on radiomics: A systematic review and meta-analysis
Background Epidermal growth factor receptor tyrosine kinase inhibitors (EGFR-TKIs) have revolutionized the prognosis for patients with EGFR-mutant lung cancer. The emergence of the T790M resistance mutation compromises the efficacy of EGFR-TKI therapy. Therefore, assessing EGFR T790M mutation status during non-small cell lung cancer (NSCLC) treatment is crucial for improving NSCLC prognosis. Method PubMed, Embase, Web of Science databases, China National Knowledge Infrastructure, and Wanfang as primary sources were systematically searched up to January 1, 2026. To assess the risk of bias and study quality, we employed the Quality Assessment of Diagnostic Accuracy Studies (QUADAS) tool and the Radiomics Quality Score version 2.0 (RQS). The diagnostic accuracy of radiomics for detecting T790M in NSCLC patients was evaluated by calculating the area under the curve (AUC), sensitivity, specificity, and accuracy for each study. Results This meta-analysis analyzed 13 studies with 2,654 patients. The pooled AUC, sensitivity, and specificity of internal validation models were 0.91, 0.73, and 0.95, respectively. The pooled AUC, sensitivity, and specificity of external validation models were 0.81, 0.73, and 0.87, respectively. Subgroup analysis revealed that imaging examinations derived from lung and mediastinal metastases achieved the highest sensitivity (0.76; 95% CI, 0.73–0.79), whereas those based on brain metastases exhibited the highest specificity (0.95; 95% CI, 0.95–0.96). The high specificity of the lung/mediastinal models was further confirmed in external validation (0.96; 95% CI, 0.95–0.98). Compared with CT, MRI-based models demonstrated a trade-off in internal validation: lower sensitivity (0.72 vs. 0.75) but significantly higher specificity (0.96 vs. 0.80). Notably, in external validation, CT achieved superior sensitivity (0.96, 95% CI 0.94–0.99). ITK-SNAP demonstrated higher sensitivity (internal: 0.76 [95% CI, 0.73–0.79]; external: 0.76 [95% CI, 0.67–0.84]) and lower specificity (internal: 0.80 [95% CI, 0.76–0.85]; external: 0.83 [95% CI, 0.70–0.95]). When stratified by a median RQS exceeding 20, higher-scoring studies were associated with higher pooled sensitivity (0.76 [95% CI, 0.70–0.82]) but a lower specificity (0.85 [95% CI, 0.79–0.90]). While in external validation, RQS ≤ 20 demonstrated higher sensitivity (0.75 [95% CI, 0.68–0.82], P < 0.001). Integrating clinical factors with radiomics improved sensitivity but reduced specificity compared with radiomics-only models (0.79 vs. 0.72 and 0.81 vs. 0.95, respectively). A similar sensitivity-specificity trade-off was observed with standardized data processing (sensitivity: 0.76 vs. 0.72; specificity: 0.80 vs 0.95). Conclusion Radiomics, as a non-invasive detection method, has demonstrated significant potential in predicting the T790M mutation status in NSCLC, showing promising clinical application prospects based on retrospective evidence. However, further standardization and validation are required in future studies. Systematic review registration https://www.crd.york.ac.uk/PROSPERO/view/CRD420251130164 (CRD420251130164).
Cloud attenuation estimation based on cloud particle sensor sonde observations for next-generation satellite communications
Abstract Cloud attenuation due to cloud particles can be a critical issue for next-generation sub-terahertz (sub-THz), terahertz (THz), and free-space optical (FSO) satellite communications. Accurate estimation of cloud attenuation under various cloud conditions is essential for the design of satellite communication systems; accordingly, numerous cloud attenuation estimation models have been proposed. The widely used ITU-R model proposed up to 200 GHz is based on liquid water content and does not explicitly consider ice crystals. Previous studies have often assumed water droplets to be dominant, potentially leading to unrealistic cloud attenuation estimates, especially in multilayered cloud systems where ice crystals can exist over a wide temperature range. In addition, approaches based on reanalysis data and numerical weather prediction models rely on parameterized cloud representations and may involve substantial uncertainties. These limitations highlight the need for observationally grounded evaluations that explicitly account for both water droplets and ice crystals. In this study, cloud attenuation was estimated up to wavelengths of approximately 20 $$\mu \textrm{m}$$ (corresponding to about 10 THz) using ten cloud microphysical data obtained from cloud particle sensor (CPS) sondes observed in Okinawa, Japan, during the Baiu (East Asia rainy season) in 2016, 2017, and 2025. CPS sondes enable the vertical measurement of the number of cloud particles along the sonde flight path and allow for their rough classification into water droplets and ice crystals, making them particularly well-suited to addressing the aforementioned challenges. Cloud attenuation was estimated based on in situ observational data obtained using CPS sondes. To the best of our knowledge, this study is the first to estimate cloud attenuation for satellite communications using in situ observational data of cloud particles. The results demonstrate that cloud attenuation due to ice crystals becomes dominant, particularly at wavelengths shorter than 1.0 mm, challenging the conventional picture that cloud attenuation is dominated by water droplets. This finding was obtained by explicitly considering the size distribution of ice crystals. A key contribution of this study is the identification of limitations in conventional cloud attenuation models for satellite communications using in situ cloud particle observations. Expanding similar studies globally could lead to the development of more accurate and widely applicable cloud attenuation models.
Cattle and human organoids reveal 2.3.4.4b H5N1 cross-species transmission potential and neuraminidase-specific neutralizing antibodies in humans
Abstract The unexpected circulation of clade 2.3.4.4b H5N1 influenza viruses in dairy cattle and the transmission to diverse mammalian species poses a pandemic risk. We sought to explore cattle and human respiratory susceptibility to the 2.3.4.4b H5N1 virus. We establish long-term expandable cattle airway and mammary organoids. The 2.3.4.4b H5N1 virus exhibits high replicative fitness in cattle mammary organoids, recapitulating its remarkable mammary tropism. The virus also replicates robustly in cattle airway organoids, suggesting an underrecognized respiratory component in ongoing outbreaks. Interestingly, human airway and nasal organoids are highly susceptible to the 2.3.4.4b H5N1 virus. Yet, a novel organoid-based neutralization assay reveals that N1 antibodies in human sera had cross-neutralizing activity against the 2.3.4.4b H5N1 and ancestral H5N1-VN1194 viruses. The cross-neutralization, exclusively manifested in the organoid-based assay, is enhanced after seasonal influenza vaccination and diminished after depleting N1-specific antibodies. Therefore, cross-neutralizing N1 antibodies are likely limiting zoonotic infection by H5N1 viruses in humans.
Why we need dedicated insect microphones - A comparison between measurement and MEMS microphone arrays highlights gap in available hardware
Acoustic insect recognition could potentially become a key element of insect sensing devices for various applications in pest management and ecological surveying. However, research in this field has yet to bridge the gap between successful laboratory studies and real-world applications. This study reports on extensive experiments and simulations carried out that aim to quantify the influence of different microphone qualities on the prospects of acoustic insect recognition systems that use them. The first microphone used is a four-channel, high-end, low-noise measurement microphone array that represents the cutting edge of acoustic recording capabilities. The second microphone is a low-budget six-channel micro-electro-mechanical system (MEMS) microphone array, initially designed for home voice assistant applications. The study clearly shows that higher-sensitivity microphones will allow for the recording and recognition of much smaller and quieter insects. However, even the simple MEMS microphones can be used to successfully recognise some insects, even in simulated real-world noisy conditions.
Genetic and molecular signatures highlight diverse pathways linking obesity to type 2 diabetes
Abstract Obesity is a major risk factor for type 2 diabetes, a disease affecting approximately 10% of the global population. Obesity is a heterogeneous condition, with different components exerting distinct, and sometimes opposing, effects on type 2 diabetes risk. Here, we aim to identify molecular mechanisms through which distinct components of obesity influence risk for type 2 diabetes by integrating multi-omics data. We identify SNPs associated with body mass index in males and females and cluster them based on their Mendelian randomisation estimates on type 2 diabetes risk. This analysis reveals four SNP clusters with distinct effects on type 2 diabetes ranging from strongly harmful to protective. We perform cluster-specific two-sample Mendelian randomisation analyses across over 3000 molecular traits to delineate metabolic, lipid, endocrine and glycaemic pathways mediating the harmful and protective effects of distinct obesity components on type 2 diabetes risk.
Retraction: Unlocking employee creativity: How learning orientation and transformational leadership spark innovation through creative self-efficacy
Assessing hypochlorite selectivity of corrosion resistance catalysts for alkaline seawater splitting
Abstract Using seawater for hydrogen production via electrolysis is increasingly attractive for cost savings. The main challenge is the competition at the anode between hydroxide oxidation and chloride oxidation, which must be minimised in both acidic and alkaline environments. Strategies such as electrostatic repulsion and ion selectivity have been used to reduce chloride oxidation in seawater, but no direct comparison has determined which is more effective. Here, we show two strategies using a rotating ring disc electrode setup to detect hypochlorite formation at the disc in an alkaline saline solution with different catalysts. We show that adding chromium to an oxygen evolution reaction catalyst reduces hypochlorite formation from 2.362 mM to 0.670 mM over 5 minutes. The electrostatic repulsion strategy (using sulphide doping) lowers the concentration to 0.966 mM under the same conditions, indicating that the ion selectivity strategy is more effective at reducing hypochlorite in alkaline seawater. However, the electrostatic repulsion strategy causes only a minor loss in electrochemical performance but improves selectivity by reducing hypochlorite formation. This research could offer insights into streamlining the design of oxygen evolution reaction catalysts and enhancing the efficiency of seawater electrolysis.
Retraction: Enhancing the decision optimization of interaction design in sustainable healthcare with improved artificial bee colony algorithm and generative artificial intelligence
Hierarchical optimization predicts plasticity in the macaque inferior temporal cortex following object training
Abstract How does the primate brain coordinate plasticity when learning to discriminate new objects? We measured consequences of object learning on inferior temporal (IT) cortex, a key waypoint supporting object recognition in the ventral visual stream, in male macaques. Neural activity in task-trained monkeys’ IT showed increased object selectivity, enhanced linear separability across objects, and more object-invariant representations compared to task-naïve monkeys. To model these differences, we developed a computational framework using anatomically-mapped artificial neural network (ANN) models of the ventral stream with various learning algorithms. Simulations revealed that gradient-based, performance-optimizing updates of ANN internal representations accurately approximated observed IT cortex changes. These models predict novel training-induced phenomena in the IT cortex, including changes independent of object identity and IT’s alignment with behavior. This convergence between empirical measurements and model predictions suggests ventral stream plasticity follows task optimization principles well-approximated by gradient descent, enabling accurate predictions about visual plasticity and generalization to test images.
Archaeomagnetic dating as a tool to overcome the Hallstatt plateau: A combined chronological approach at the salt production site of Piscina Torta (Rome, Italy)
The salt production site of Piscina Torta (Rome, Italy) presents a complex chronological challenge due to its ceramic assemblage, which is almost exclusively represented by coarse ware jars with limited morphological variability. Furthermore, its occupation period coincides with the Hallstatt plateau, which hampers radiocarbon dating resolution. This study addresses these limitations by applying archaeomagnetic analyses to a kiln and ceramic fragments both from Area 1 of the site. Archaeomagnetic dating based on directions from the kiln and on archaeointensity from pottery sherds yielded two-time intervals: 930–755 BCE and 740–700 BCE. These results confirm the potential of this method as a viable dating tool, especially for problematic radiocarbon periods. At Piscina Torta, the integration of archaeomagnetic, typo- chronological, and radiocarbon data enabled the subdivision of Area 1 into three occupational phases: Phase A (725–710 BCE), Phase B (710–550 BCE), and Phase C (550–525 BCE). This study demonstrates the effectiveness of archaeomagnetism in enhancing chronological resolution in early first-millennium BCE contexts and highlights its important contribution when combined with traditional dating methods.
Body surface potential mapping of the cortico-muscular axis using smart textile electrode arrays
Abstract Cutaneous electrophysiology is a fundamental non-invasive technique for assessing electrically active organs such as the brain, heart, and muscles. Standard approaches, however, are limited in spatial resolution, reducing sensitivity to certain pathological features. The development of body surface potential mapping using electrode arrays has helped overcome these limitations, enhancing the diagnostic power of cutaneous recordings, yet clinical adoption remains constrained by challenges in electrode performance, wiring complexity, wearability, data transmission, and interpretability. Here, we present a hybrid e-textile electrode array system that overcomes these barriers, enabling simultaneous mapping of electrical activity along the cortico-muscular axis. The system combines application-specific conducting polymer coatings to improve electrode performance, a flexible fabrication process for robust connectivity and wearability, and interpretable machine learning algorithms for data analysis. In controlled single-subject experiments, we demonstrate reliable muscle and brain recordings, enabling classification of grasped object shapes and somatosensory stimuli. Simultaneous multi-site recordings along the cortico-muscular axis provide spatial maps of reaction time distributions and allow prediction of muscle activation patterns from cortical activity. This platform establishes a framework for wearable, multi-modal electrophysiological mapping and non-invasive study of cortico-muscular dynamics, representing a step towards practical brain–body interfaces with applications in neurorehabilitation, prosthetics, and human–machine interaction.
Retraction: Optimal performance selection of sustainable mobility service projects based on IFSS ‐ Prospect theory ‐ VIKOR: A case study of electric vehicle sharing program
High-performance stretchable conductive film on complex microscale structures via spontaneous fusion of liquid metal
Value dynamics and interpersonal tension among astronauts during International Space Station missions
Astronauts’ ability to maintain motivation over extended periods is crucial for space mission success. This paper examines how motivational goals (personal values) change during space missions and explores the associations between perceived value congruence and intra-crew tension among astronauts staying 4–7 months at the International Space Station. Twelve astronauts regularly completed the Crew Values Questionnaire (CVQ) package, assessing personal values, perceived value congruence of crew members, and tension attributed to perceived value incongruence. Overall, they rated their values, particularly universalism, benevolence, and tradition, as highly concordant. Temporal analyses showed that scores for hedonism and power increased early in missions and declined later, while benevolence and security decreased and rose again towards the end and into post-mission. Perceived value-congruence followed similar trajectories, with differences in achievement, power, and benevolence decreasing and then increasing across mission phases. Multilevel modeling showed that perceived incongruence in seven out of eight personal values significantly predicted interpersonal tension. In conclusion, in-mission adjustments of value priorities may have helped astronauts sustain motivation, but these shifts could also influence crew dynamics. Pre-mission training and in-flight support should target shifts in motivational sources and manage interpersonal tensions from value diversity to prevent adverse outcomes and leverage crew heterogeneity.