Browse Articles
Discover research articles across all indexed journals
The impact of Streptococcus thermophilus IDCC 2201 on gut microbiota and its potential as a prophylactic agent for colorectal cancer
Abstract Colorectal cancer (CRC) is one of the leading causes of cancer-related deaths. The gut microbiota is known to be closely associated with CRC development, interacting with each other and influencing host metabolism. Probiotic interventions have been applied to prevent CRC; however, the mechanisms underlying their effects on CRC remain unclear. In this study, we evaluated the effects of the probiotic strain Streptococcus thermophilus IDCC 2201 on its interactions with gut commensal bacteria and CRC cell viability. S. thermophilus inhibited the growth of most bacterial species comprising the human gut microbiota, with the exception of Bacteroides dorei . Further studies identified several factors produced by S. thermophilus , such as xylan-degrading enzymes and B vitamins, which promoted B. dorei growth through nutrient cross-feeding. During co-culture of S. thermophilus with individual gut commensals, bioactive compounds such as folate were significantly produced, further inhibiting CRC cell viability along with the culture supernatant of S. thermophilus . These findings suggest that S. thermophilus IDCC 2201 may serve as a potential prophylactic agent for colon cancer, with its effects mediated through interactions with gut commensal bacteria.
Cycle based state of health estimation of lithium ion cells using deep learning architectures
Abstract State of Health estimation in lithium-ion batteries is critical for reliable operation in electric vehicles and energy storage systems. This work evaluates four deep learning models—Multilayer Perceptron, Gated Recurrent Unit, Long Short-Term Memory, and Temporal Convolutional Network for cycle-based SoH prediction using discharge data from the NASA B0005, B0006, and B0007 cells. SoH values were obtained by numerical integration of discharge current and normalized with respect to the initial capacity. All models were implemented in PyTorch and assessed using RMSE, MAE, and R² metrics. On B0005, the MLP achieved RMSE 0.0069, MAE 0.0049, and R² = 0.9955, with TCN showing similar accuracy. Results on B0006 and B0007 confirmed the stability of MLP and TCN predictions across different cells. Residuals remained tightly clustered, and loss curves indicated smooth convergence. GRU and LSTM required higher training time without accuracy improvements. MLP demonstrated the best balance of accuracy and computational efficiency, making it suitable for embedded battery management systems. TCN provided robust accuracy with moderate complexity. The results verify that data-driven deep learning methods can capture nonlinear degradation behavior consistently across multiple cells.
Does gravity produce quantum weirdness? Proposal divides physicists
UAV-asisted IoT network framework with hybrid deep reinforcement and federated learning
Daily briefing: People with cancer lived longer if they’d had a COVID-19 vaccine
Structural influence on titanium ion dissolution in 3D-printed Ti6Al4V orthopedic implants
Abstract 3D printed orthopedic implants have emerged as innovative solutions for treating bone tumors, offering advantages such as patient-specific customization and faster production compared to conventional manufacturing methods. However, elevated concentrations of titanium (Ti) ions in the bloodstream have frequently been observed following limb salvage surgery using 3D printed Ti6Al4V implants, which could lead to systemic toxicity and critical implant failure. In this study, we characterize the Ti dissolution phenomenon associated with 3D printed implants. Finite element analysis (FEA) of full-scale pelvic and tibial implants revealed that large mesh surface areas designed for implant–tissue integration can accelerate corrosion. Microstructural analyses of cubical Ti6Al4V samples with solid, mesh, and solid-mesh hybrid geometries revealed that galvanic coupling between the alpha (α) and beta (β) phases drives localized corrosion. A notable difference in β-phase content—ranging from 145% to 200%—was observed among the three cases, with the highest β-phase content in the mesh structures. These findings indicate that although mesh structures are essential for implant–tissue bonding, they can significantly promote Ti ion release, potentially compromising the mechanical integrity of the implant over time. Careful design and surface treatment strategies are therefore needed to balance biological integration with long-term material stability.
Mapping residual water fluoride levels and geological patterns in a rural semi-arid area of Brazil
Partial orientation retrieval of proteins from X-ray free-electron-laser induced explosions for single particle imaging
Abstract Single Particle Imaging techniques at X-ray lasers have made significant strides, yet the challenge of determining the orientation of freely rotating molecules during delivery remains. In this study, we propose a novel approach to partially retrieve the relative orientation of proteins exposed to ultrafast X-ray pulses by analyzing the fragmentation patterns resulting from Coulomb explosions. We simulate these explosions for 85 proteins in the size range 100 – 4000 atoms using a hybrid Monte Carlo/Molecular Dynamics approach and capture the resulting ion ejection patterns on two virtual detectors. We exploit information from the explosion to infer orientations of proteins at the time of X-ray exposure. Our results demonstrate that partial orientation information can be extracted, particularly for larger proteins. We conclude that knowledge on ion data from X-ray laser induced explosions can directly provide the sample’s relative orientation, complementary to traditional orientation-retrieval algorithms based on diffraction patterns.
Uncertain climate effects of anthropogenic reactive nitrogen
Glutaminase inhibitor CB-839 causes metabolic adjustments in colorectal cancer cells
Classical theories of gravity produce entanglement
Abstract The unification of gravity and quantum mechanics remains one of the most profound open questions in science. With recent advances in quantum technology, an experimental idea first proposed by Richard Feynman 1 is now regarded as a promising route to testing this unification for the first time. The experiment involves placing a massive object in a quantum superposition of two locations and letting it gravitationally interact with another mass. If the two objects subsequently become entangled, this is considered unambiguous evidence that gravity obeys the laws of quantum mechanics. This conclusion derives from theorems that treat a classical gravitational interaction as a local interaction capable of transmitting only classical, not quantum, information 2–8 . Here we extend the description of matter used in these theorems to the full framework of quantum field theory, finding that theories with classical gravity can then transmit quantum information and, thus, generate entanglement through physical, local processes. The effect scales differently to that predicted by theories of quantum gravity, and so it gives information on the parameters and form of the experiment required to robustly provide evidence for the quantum nature of gravity.
The spatial heterogeneity of soil physical and chemical properties and comprehensive fertility in the Leizhou Peninsula coast
Copper-free chips generate robust ‘combs’ of multicoloured light
Characterizing volatile organic compound profiles in oral cancer using multiple sample collection approaches by GC-IMS and TD-GC-MS
A comprehensive benchmark of active learning strategies with AutoML for small-sample regression in materials science
Abstract The high cost and difficulty of acquiring labeled data in materials science often limits the scale of data-driven modeling efforts. Experimental synthesis and characterization often require expert knowledge, expensive equipment, and time-consuming procedures, making it critical to develop data-efficient learning strategies. Integrating Automated Machine Learning (AutoML) with active learning enables the construction of robust material-property prediction models while substantially reducing the volume of labeled data required. This benchmark study aims to evaluate various active learning (AL) strategies within AutoML in materials science regression tasks. The performance of each strategy in terms of model accuracy and data efficiency is analyzed. The 9 datasets used are derived from materials formulation design, which are typically small due to high data acquisition costs. 17 active-learning strategies, together with a Random-Sampling baseline, are systematically evaluated and compared for their effectiveness. Early in the acquisition process, uncertainty-driven (LCMD, Tree-based-R) and diversity-hybrid (RD-GS) strategies clearly outperform geometry-only heuristics (GSx, EGAL) and baseline, selecting more informative samples and improving model accuracy. As the labeled set grows, the gap narrows and all 17 methods converge, indicating diminishing returns from AL under AutoML.
Gut microbiota composition combined with reduced intestinal fatty acid uptake prevents hepatic steatosis in obesity-resistant mice fed a high-fat diet
Biological rhythm and depression severity mediate the relationship between adverse childhood experiences and non-suicidal self-injury among adolescents with depressive disorder
Joint neutrino oscillation analysis from the T2K and NOvA experiments
Abstract The landmark discovery that neutrinos have mass and can change type (or flavour) as they propagate—a process called neutrino oscillation 1–6 —has opened up a rich array of theoretical and experimental questions being actively pursued today. Neutrino oscillation remains the most powerful experimental tool for addressing many of these questions, including whether neutrinos violate charge-parity (CP) symmetry, which has possible connections to the unexplained preponderance of matter over antimatter in the Universe 7–11 . Oscillation measurements also probe the mass-squared differences between the different neutrino mass states (Δ m 2 ), whether there are two light states and a heavier one (normal ordering) or vice versa (inverted ordering), and the structure of neutrino mass and flavour mixing 12 . Here we carry out the first joint analysis of datasets from NOvA 13 and T2K 14 , the two currently operating long-baseline neutrino oscillation experiments (hundreds of kilometres of neutrino travel distance), taking advantage of our complementary experimental designs and setting new constraints on several neutrino sector parameters. This analysis provides new precision on the $$\Delta {m}_{32}^{2}$$ Δ m 32 2 mass difference, finding $$2.4{3}_{-0.03}^{+0.04}\times 1{0}^{-3}\,{{\rm{eV}}}^{2}$$ 2.4 3 − 0.03 + 0.04 × 1 0 − 3 eV 2 in the normal ordering and $$-2.4{8}_{-0.04}^{+0.03}\times 1{0}^{-3}\,{{\rm{eV}}}^{2}$$ − 2.4 8 − 0.04 + 0.03 × 1 0 − 3 eV 2 in the inverted ordering, as well as a 3 σ interval on δ CP of [−1.38π, 0.30π] in the normal ordering and [−0.92π, −0.04π] in the inverted ordering. The data show no strong preference for either mass ordering, but notably, if inverted ordering were assumed true within the three-flavour mixing model, then our results would provide evidence of CP symmetry violation in the lepton sector.