Search Results (82)

This research introduces a new spectral pruning approach using the Adjacency Edge Index (AEI), which is a centrality measure from the theory of spectral graphs and first-order matrix perturbation theory. The AEI score reflects the contribution of each neuron to the network’s dynamic synchronizability via the Fiedler vector. The AEI approach thus offers a mathematically motivated saliency score in the context of data-driven neuron co-activation graphs. The proposed approach has been tested on MNIST, Fashion MNIST, KMNIST, and a real-world social network dataset. The approach has been extended to convolutional filter pruning on the CIFAR-10 dataset using spatial global average pooling. The AEI approach has been compared to magnitude-based pruning methods like L1 and L2 norms and gradient-based pruning methods like SNIP and GraSP. The robustness of the proposed approach has been established by comparing the results over five random seeds. The AEI approach is proposed as a principled, interpretable, structure-aware pruning criterion rather than an accuracy-maximising method. Spectral analysis demonstrates that AEI is the only evaluated method that systematically targets structurally peripheral neurons, whereas magnitude-based methods prune broadly across the structural spectrum and GraSP actively removes structurally central neurons, an effect most pronounced on more complex datasets and deeper architectures (CIFAR-100, ResNet-20), where it causes substantial accuracy degradation at high sparsity. The AEI approach has been extended to the Hybrid approach by combining the AEI and L2 norms. The Hybrid approach has been seen to improve the accuracy gap at 40% sparsity from 5.46 to 2.26 percentage points over the state of the art. The robustness of the proposed approach has been established by conducting ablation studies on the robustness of the approach to the selection of the graph. The approach has been seen to be moderately robust to the selection of the graph with Spearman’s $\rho$ ≈ 0.70. The accuracy gap has been seen to be less than 0.2%. Full article

Natural products featuring a direct σ-bond between a cyclic monoterpene and an aromatic moiety provide a vast source of biological activities, such as antimicrobial, anticancer, antiviral, anticoagulant and cannabinoid regulatory, among others. Only few methods exist for synthetically accessing such hybrid structures and their analogs, all of which are prone to limitations, most notably the reliance on sensitive organometallic intermediates and the difficulty in furnishing certain stereoisomers. An efficient, three-stage synthetic methodology is described herein, that enables the production of hybrid structures featuring a C(sp³)-C(sp²) bond between six-membered cyclic monoterpenes and aromatic moieties. This process combines: enol triflate formation from a terpenoid ketone precursor, that introduces most of the stereochemical information; Suzuki–Miyaura C-C cross-coupling of the enol triflate with a pool of (hetero)arylboronic acids, to establish the terpene–aromatic link, initially in the form of a C(sp²)-C(sp²) bond; and a stereoselective hydrogenation of the resulting adducts to afford the target compounds, establishing the stereoconfiguration of the last chiral center. Enantiomeric terpenoid scaffolds derived from menthone and trans-tetrahydrocarvone have been combined with six (6) (hetero)arylboronic acids, including medicinally relevant moieties, such as methoxyphenyl, pyridine, quinoline and benzofuran. The power of this method, apart from circumventing the need for in situ-formed sensitive organometallic intermediates, resides in providing access, for the first time, to menthyl- and trans-tetrahydrocarvoneyl-type stereoisomers, that were unattainable by any previously described method. The resulting compound library members exhibit drug-like features, based on the computational assessment of 11 selected physicochemical parameters (molecular weight, polarity, aqueous solubility, degree of unsaturation, conformational flexibility, lipophilicity, BBB permeability, skin permeability, gastrointestinal absorption, P-glycoprotein substrate behavior and Lipinski compatibility), using the platforms SwissADME, ADMETLab 3.0 and pkCSM. A computational docking study employing AutoDock Vina further identified promising candidates for targeting the known binding sites of human cannabinoid receptors CB1 and CB2, with calculated binding affinities comparable to those of established ligands. Full article

Developing lead-free organic-inorganic hybrid metal halides with strong light-harvesting capabilities and superior stability, while precisely tuning their crystalline phases and electronic structures, remains a key challenge in optoelectronics. Herein, we report a neutral iodine-induced structural transition from a 1D chain-like (C₆H₁₁NH₃)₂SbI₅ architecture to a 0D dimeric (C₆H₁₁NH₃)₃[Sb₂I₉]·I₂ supramolecular host-guest complex. This transformation is achieved via a controlled solution-cooling crystallization process, yielding high-quality bulk single crystals. Crystallographic analysis reveals that N–H···I hydrogen-bonding networks stabilize the organic cations, while halogen bonding interactions anchor the I₂ guests within the lattice cavities of the [Sb₂I₉]³⁻ dimeric host. Experimental characterizations, including XRD, TGA, and XPS, confirm the high phase purity and thermal stability of the (C₆H₁₁NH₃)₃[Sb₂I₉]·I₂ hybrid and determine its electronic band structure. To further elucidate the underlying mechanisms, theoretical calculations were performed, revealing that strong sp-orbital hybridization yields a high absorption coefficient. The associated dimensional transition narrows the direct optical bandgap to 1.46 eV, approaching the Shockley-Queisser limit and demonstrating strong potential for visible-light harvesting. This work elucidates the role of supramolecular host-guest interactions in modulating the lattice evolution of lead-free antimony-based materials, presenting halogen guest engineering as an effective approach for optoelectronic material design. Full article

The growing need for secure image transmission across public networks requires robust encryption algorithms. Traditional chaos-based image ciphers typically have a small key space, weak avalanche behavior, or are susceptible to differential cryptanalysis. To overcome such inadequacies, this paper suggests a new adaptive image cryptosystem that combines a fractional-order memristive chaotic engine and a non-linear hybrid encryption kernel. The system uses piano-inspired feedback; the keystream generator dynamically adapts to the previously encrypted pixel, enabling powerful Cipher Block Chaining (CBC)-style chaining and content-dependent diffusion. A four-dimensional memristive system is solved by the use of fractional-order calculus, which gives an ultra-large key space (>10⁸⁰) and very high sensitivity to initial conditions—confirmed by a positive largest Lyapunov exponent (1.7199). The encryption kernel maps the traditional Exclusive OR (XOR) with the reversible two-step operation: the modular addition of the plaintext with the first keystream byte and the XOR with the second keystream one, both of which increase non-linearity and confusion. Large-scale experiments with six standard 256 × 256 colour images indicate almost ideal entropy (7.9994), Number of Pixel Change Rate (NPCR) which is 99.62, Unified Average Changing Intensity (UACI) which is 33.43, correlation coefficients are near to zero, very low Gray-Level Co-occurrence Matrix (GLCM) homogeneity (≈0.017) and high contrast (≈4843) and low energy (≈0.006 The ciphertext passes seven National Institute of Standards and Technology (NIST) SP-800-22 statistical tests, is extremely sensitive to keys (a perturbation of 1 × 10⁻¹⁴ alters >99.6% of ciphertext) and resists chosen-plaintext and known-plaintext attacks. Decryption has linear time complexity O(N), and average encryption and decryption times are 3.40 s and 2.75 s for 256 × 256 images. The proposed cryptosystem provides an attractive security–performance trade-off that can be used in high-security systems like medical image protection, privacy-preserving multimedia transmission, and secure cloud storage. Full article

Endophytic bacteria from plants adapted to arid and semi-arid environments represent an underexplored reservoir of microbial diversity with potential agricultural applications. Here, we report a polyphasic taxonomic and genome-based characterization of Achromobacter sp. isolates recovered from root and foliar tissues of Citrullus colocynthis and Peganum harmala, two medicinal plants thriving under harsh environmental conditions. Whole-genome sequencing, phylogenomic analyses, average nucleotide identity (ANI), digital DNA–DNA hybridization (dDDH), multilocus sequence typing, and detailed phenotypic profiling revealed three previously undescribed species, for which we propose the names Achromobacter colocynthi sp. nov., Achromobacter maghribensis sp. nov., and Achromobacter semiaridum sp. nov. Genome assemblies were highly complete (98.7–99.2%) with minimal contamination (<1%), supporting robust taxonomic inference. All three species displayed ANI and dDDH values below accepted thresholds relative to their closest phylogenetic neighbors, despite partial inconsistencies in 16S rRNA similarity for one isolate, highlighting the value of genome-wide metrics for species delineation. Phylogenomic analyses placed the novel taxa within Achromobacter sp. as distinct evolutionary lineages. Phenotypic characterization indicated broad metabolic versatility, including utilization of carbohydrates, organic acids, and amino acids, tolerance to moderate salinity and acidic pH, and resistance to multiple antimicrobial compounds, traits likely linked to adaptation to endophytic lifestyles under semi-arid conditions. Beyond their taxonomic novelty, the isolates exhibited in vitro traits associated with plant adaptation and stress tolerance, including IAA production, ACC deaminase activity, and tolerance to Zn, Cu, and Cd. Genomic analyses further indicated functions related to phosphate acquisition and stress response. These findings expand the taxonomic framework of Achromobacter sp., establish C. colocynthis and P. harmala as reservoirs of novel endophytic bacteria, and highlight their potential relevance for agricultural biotechnology in stress-prone environments. Full article

This study addresses the challenge of high signal-to-noise ratios in financial sentiment analysis by introducing a hybrid, multi-stage AI framework. We combine the high-throughput capabilities of FinBERT with the deep contextual reasoning of Google Gemini to extract actionable intelligence from over 9,000,000 data points, including the U.S. Securities and Exchange Commission (SEC) filings and financial news. By applying our rigorous “Data Funnel” logic, we filter out noise from the massive dataset and surface a small set of high-conviction signals. These signals are executed on a historically dynamic universe of top S&P 500 constituents within a dollar-neutral long/short framework, integrated with macro-regime filters and technical trend confirmation. Our results over a 16-year testing period demonstrate a mean excess return of 51.02% per annum net of transaction costs, while achieving a Sharpe ratio of 1.06 and a Sortino ratio of 2.61. The significant divergence between Sharpe and Sortino ratios highlights the strategy’s positive skewness, effectively capturing upside volatility while limiting downside risk. Statistical robustness is confirmed by a Newey–West adjusted t-statistic of 4.01, indicating that the generated alpha is highly significant. This research provides a proof-of-concept for the use of Large Language Models (LLMs) as qualitative gatekeepers in quantitative finance, effectively bridging the gap between statistical NLP and human-like contextual understanding. Full article

Chinese tongue sole (Cynoglossus semilaevis) is an important aquaculture species in China. Under high-temperature conditions, genetically female fish can undergo sex reversal and develop into phenotypic males (pseudomale fish). Previous studies have demonstrated that apoptosis might function in sex differentiation. Based on this, we identified and characterized the functions of six caspase genes (caspase1-like, caspase3a, caspase6, caspase8, caspase8-like, caspase9) in Chinese tongue sole. These six caspase genes were expressed in all analyzed tissues of both males and females. They were detected to be expressed in the gonads at various developmental stages, with expression levels peaking between 7 months and 2 years of age. In situ hybridization (ISH) analysis showed that the caspase genes were mainly localized in spermatocytes and oocytes. Promoter activity analysis indicated that with the exception of caspase3a, the remaining five caspase genes exhibited promoter activity and were regulated by transcription factors, including sp1 and gata4. High-temperature stimulation can significantly affect the expression of caspases in the gonads of both male and female fish, with female fish showing a more pronounced response. An siRNA-mediated knockdown experiment revealed that following caspase knockdown, the expression of sex differentiation-related genes, heat shock transcription factors (hsf), and heat shock proteins (hsp) in Chinese tongue sole was significantly altered. Based on these findings, we speculate that caspases play an important role in the sex differentiation process by responding to temperature stimuli. Full article

Stock price forecasting remains challenging due to the non-linear, noisy, and non-stationary nature of financial time series. Although LSTMs and Transformer-based models have improved sequential modeling, their ability to scale efficiently to long financial sequences remains limited. Recently, selective state space models such as Mamba have emerged as efficient alternatives to self-attention, offering attention-like performance with linear computational complexity. In this study, we systematically evaluate multiple Mamba-augmented Transformer architectures for stock market price forecasting. We further propose CrossMamba, a novel architecture that models cross-sequence interactions between encoder and decoder representations using a causal Mamba block. Experiments on multiple S&P 500 and Yahoo Finance stocks show that CrossMamba achieves superior short-horizon performance with 5-day input windows (${\mathit{R}}^2$ up to 0.963), while Hybrid Bi-Mamba performs best for longer horizons, achieving the lowest MAE of 0.67 for 10-day forecasts. Compared with advanced Mamba-based and Transformer baselines, the proposed models achieve competitive accuracy while maintaining substantially improved computational efficiency. These results highlight the effectiveness of Mamba-augmented Transformers as scalable architectures for financial time series forecasting. Full article

This paper presents Nexus, a proof-of-concept low-cost, modular, and reprogrammable multichannel data logger aimed at validating the architectural feasibility of an open and scalable acquisition platform for scientific instrumentation. The system was conceived to address common limitations of commercial data loggers, such as high cost, restricted configurability, and limited autonomy, by relying exclusively on widely available components and open hardware/software resources, thereby facilitating reproducibility and adoption in resource-constrained academic and industrial environments. The proposed architecture supports up to six interchangeable acquisition modules, enabling the integration of up to 20 analog channels with heterogeneous resolutions (24-bit, 12-bit, and 10-bit ADCs), as well as digital acquisition through multiple communication interfaces, including I²C (two independent buses), SPI (two buses), and UART (three interfaces). Quantitative validation was performed using representative acquisition configurations, including a 24-bit ADS1256 stage operating at sampling rates of up to 30 kSPS, 12-bit microcontroller-based stages operating at approximately 1 kSPS, and 10-bit operating at 100 SPS, consistent with stable real-time acquisition and visualization under proof-of-concept constraints. SPI communication was configured with an effective clock frequency of 2 MHz, ensuring deterministic data transfer across the tested acquisition modules. A hybrid data management strategy is implemented, combining high-capacity local storage via USB 3.0 solid-state drives, optional cloud synchronization, and a 7-inch touchscreen human–machine interface based on Raspberry Pi OS for system control and visualization. Power continuity is addressed through an integrated smart uninterruptible power supply, which provides telemetry, automatic source switching, and limited backup operation during power interruptions. As a proof of concept, the system was functionally validated through architectural and interface-level tests, demonstrating stable communication across all supported protocols and reliable acquisition of synthetic and biosignal-like waveforms. The results confirm the feasibility of the proposed modular architecture and its ability to integrate heterogeneous acquisition, storage, and interface subsystems within a unified open-source platform. While not intended as a finalized commercial product, Nexus establishes a validated foundation for future developments in modular data logging, embedded intelligence, and application-specific instrumentation. Full article

This study details the design, synthesis, and evaluation of a novel series of fourteen 2-((1H-1,2,4-triazol-5-yl)thio)-N-benzylidene-N-arylacetohydrazide hybrid compounds. The primary objective was to investigate their potential as antimicrobial agents and assess their cytotoxicity. A systematic approach combining in silico screening and experimental validation was employed. The initial in silico analysis, using SwissADME, identified compounds with favorable drug-like properties. Subsequently, all fourteen compounds were synthesized and characterized using various spectroscopic methods. Their antibacterial efficacy was evaluated in vitro against Gram-negative (Klebsiella aerogenes) and Gram-positive (Enterococcus sp.) bacteria through growth kinetics and colony-forming unit (CFU) assays. Cytotoxicity was assessed using MTT assays on HEK (human embryonic kidney) cell lines. The compound, 2-((1H-1,2,4-triazol-3-yl)thio)-N′-(2-fluorobenzylidene)-N-phenylacetohydrazide emerged as the most promising candidate, demonstrating broad-spectrum antibacterial activity. These findings highlight the potential of 2-((1H-1,2,4-triazol-5-yl)thio)-N-benzylidene-N-arylacetohydrazide hybrids as a scaffold for developing new antimicrobial agents. Furthermore, this study suggests possible environmental applications for these compounds in antimicrobial resistance (AMR) management. Full article

With the increasing demand for energy conservation and emission reduction in the automotive industry, optimizing the performance of cylinder liner and piston ring pairs in engines has become crucial. Aluminum–silicon alloy cylinder liners, known for their lightweight and excellent thermal conductivity, have emerged as a new trend in cylinder liner materials. Given the relatively moderate hardness of Al-Si alloys, judicious selection of piston rings is imperative to ensure optimal performance. This study investigates the tribological properties of aluminum–silicon alloy cylinder liners paired with CKS and DLC piston rings. The surface morphology and hardness of the test materials were characterized, and reciprocating friction and wear tests were conducted, using a tribometer to simulate operating conditions. The friction coefficient and wear volume were used as indicators to evaluate the tribological properties of the piston rings. The results show that, when the aluminum–silicon alloy cylinder liner was paired with a DLC piston ring, the friction coefficient was 27.82% lower, and the wear volume of the cylinder liner was 83.52% lower, compared to pairing with a CKS piston ring. When paired with a CKS piston ring, wear was exacerbated because silicon particles were easily dislodged to form abrasive particles. This particle detachment is mainly caused by the collision between the fine ceramic particles embedded in the CKS coating and the silicon particles (≤5 μm) uniformly distributed in the Al-Si alloy cylinder liner during the sliding process. The DLC piston ring, containing both sp² and sp³ hybridized carbon–carbon bonds, combined excellent lubrication properties with high hardness, resulting in minimal wear on both the cylinder liner and piston ring. Specifically, the DLC coating has a hardness of 2300 HV_0.3, which is 2.42 times that of the CKS piston ring (950 HV_0.3); the sp³-hybridized carbon in the DLC coating enhances its wear resistance to resist scratching from silicon particles in the cylinder liner, while the sp²-hybridized carbon forms a graphite-like transfer layer at the friction interface to reduce frictional resistance. In conclusion, the aluminum–silicon alloy cylinder liner paired with a DLC piston ring exhibits superior tribological properties. Selecting an appropriate piston ring can significantly enhance the tribological properties of the cylinder liner–piston ring pair, thereby extending the engine’s service life. Full article

Formaldehyde is a toxic gas commonly found in industrial emissions, and ZnO is widely used for its detection due to its excellent gas-sensing properties. Most studies focus on non-polar or low-index ZnO surfaces, whereas investigations on high-index polar surfaces remain limited. In this work, density functional theory (DFT) was employed to study CH₂O adsorption on the ZnO [$10\overline{11}$] surface. By exploring various coverages, adsorption sites, and unit cell dimensions, ten stable configurations were identified. A maximum adsorption energy of −2.19 eV/CH₂O on configuration S1 was obtained, surpassing reported low-index surfaces. Strong adsorption originated from dual unsaturated Zn bonds, which promoted C–C formation between CH₂O molecules and induced synergistic Zn–O bonding. Adsorption further led to sp³-like hybridization and O 2p/Zn 3d orbital interactions, significantly narrowing the band gap. Electron redistribution, as evidenced by charge density analysis, revealed strong electronic modulation. This work clarifies the microscopic mechanism of ZnO high-index surfaces, offering insights for optimizing gas-sensing materials. Full article

A reassessment of the risk-benefit balance of the two lipid nanoparticle (LNP)-based vaccines, Pfizer’s Comirnaty and Moderna’s Spikevax, is currently underway. While the FDA has approved updated products, their administration is recommended only for individuals aged 65 years or older and for those aged 6 months or older who have at least one underlying medical condition associated with an increased risk of severe COVID-19. Among other factors, this change in guidelines reflect an expanded spectrum and increased incidence of adverse events (AEs) and complications relative to other vaccines. Although severe AEs are relatively rare (occurring in <0.5%) in vaccinated individuals, the sheer scale of global vaccination has resulted in millions of vaccine injuries, rendering post-vaccination syndrome (PVS) both clinically significant and scientifically intriguing. Nevertheless, the cellular and molecular mechanisms of these AEs are poorly understood. To better understand the phenomenon and to identify research needs, this review aims to highlight some theoretically plausible connections between the manifestations of PVS and some unique structural properties of mRNA-LNPs. The latter include (i) ribosomal synthesis of the antigenic spike protein (SP) without natural control over mRNA translation, diversifying antigen processing and presentation; (ii) stabilization of the mRNA by multiple chemical modification, abnormally increasing translation efficiency and frameshift mutation risk; (iii) encoding for SP, a protein with multiple toxic effects; (iv) promotion of innate immune activation and mRNA transfection in off-target tissues by the LNP, leading to systemic inflammation with autoimmune phenomena; (v) short post-reconstitution stability of vaccine nanoparticles contributing to whole-body distribution and mRNA transfection; (vi) immune reactivity and immunogenicity of PEG on the LNP surface increasing the risk of complement activation with LNP disintegration and anaphylaxis; (vii) GC enrichment and double proline modifications stabilize SP mRNA and prefusion SP, respectively; and (viii) contaminations with plasmid DNA and other organic and inorganic elements entailing toxicity with cancer risk. The collateral immune anomalies considered are innate immune activation, T-cell- and antibody-mediated cytotoxicities, dissemination of pseudo virus-like hybrid exosomes, somatic hypermutation, insertion mutagenesis, frameshift mutation, and reverse transcription. Lessons from mRNA-LNP vaccine-associated AEs may guide strategies for the prediction, prevention, and treatment of AEs, while informing the design of safer next-generation mRNA vaccines and therapeutics. Full article

To address the limitations of traditional models in capturing complex features for concrete strength prediction, this study proposes a hybrid deep learning approach that integrates multiple attention mechanisms with gated recurrent units (GRU). The methodology employs a multi-scale validation framework, conducting three-dimensional validation across three datasets: the Kaggle standard dataset, the lightweight foam concrete dataset, and the self-compacting concrete dataset. Six attention mechanisms (SE attention, dot-product attention, self-attention, etc.) are comprehensively compared to optimise the GRU network structure. A Newton–Raphson-based optimiser (NRBO) enables hyperparameter adaptive tuning. Experimental results show significant improvements over the baseline GRU model: mean R² increased by 6.99%, while RMSE and MAE decreased by 38.5% and 37.5%, respectively. SHAP interpretability analysis confirms that attention mechanisms effectively capture key parameters like SP and VMA in the self-compacting concrete dataset. Based on the findings, this study recommends using self-attention for datasets smaller than 200 samples and selecting the higher-accuracy model between self-attention and stacked attention mechanisms for larger datasets. Full article

A novel Gram-negative bacterium, designated strain SN16^T, was isolated from a harmful algal bloom (HAB). Strain SN16^T exhibited potent, broad-spectrum algicidal activity against the colony-forming alga Phaeocystis globosa and eight other HAB-causing species, highlighting its potential as a promising candidate for the biological control of HABs. A phylogenetic analysis of 16S rRNA gene sequences placed strain SN16^T within the genus Flagellimonas. The average nucleotide identity (ANI) and digital DNA–DNA hybridization (dDDH) values between strain SN16^T and its relatives were 75.4–91.4% and 19.3–44.0%, respectively. These values fall below the established thresholds for species delineation, confirming that SN16^T represents a novel species. A chemotaxonomic analysis revealed its dominant cellular fatty acids to be iso-C_15:0 and iso-C_15:1 G. The major polar lipid was phosphatidylethanolamine, and the primary respiratory quinone was menaquinone-6. Genome mining identified 11 biosynthetic gene clusters (BGCs), including those encoding for terpenes, ribosomal peptide synthetases, and non-ribosomal peptide synthetases. By integrating BGC analysis with the observed algicidal activities, we predicted that pentalenolactone and xiamycin analogues are the likely causative compounds. Based on this polyphasic evidence, strain SN16^T is proposed as a novel species of the genus Flagellimonas, named Flagellimonas algicida sp. nov. This is the first report of Flagellimonas species exhibiting broad-spectrum algicidal activity, including activity against the colonial form of P. globosa—a key ecological challenge in HAB mitigation. Full article