Quantum processors encode computational information in relative phases and are therefore sensitive to spatially and temporally correlated dephasing. We formulate gravitational redshift as a physically constrained template for such noise. A fixed, known redshift produces a deterministic phase, whereas dephasing arises only when unresolved displacement, timing, or frequency fluctuations are transduced into phase uncertainty; for vertical motion, the conversion factor is βg=ω0g/c02. The gravity-specific content is thus the mapping from geometry fluctuations to phase covariance and frequency-noise spectra, while the resulting Gaussian dephasing dynamics are generic. We test spatial-kernel dependence using exponential, Gaussian, Matérn, and power-law correlations. Circuit-level effects are evaluated with an explicit GHZ-preparation circuit, showing that a full-GHZ-per-layer exponential approximation overestimates loss. For temporal correlations, we derive Sδω=βg2Sh and use a mechanically motivated drift-plus-vibration model, which attributes memory to the mechanical process rather than gravity. Platform estimates indicate that the gravity-transduced contribution is negligible compared with present technical dephasing. The framework therefore provides a falsifiable geometry-to-noise model while keeping mitigation and compilation claims quantitatively cautious.
The demand for clean water is rapidly growing, which has prompted further development of nanomaterials for environmental purification and biomedical research. Photocatalysis is a highly promising oxidation method for the removal of organic and biological pollutants. It is quite evident that there is a serious and pressing demand to design economic treatment techniques, which are cheap and energy independent. In this study, we described the fabrication of Z-scheme heterostructure Er doped ZrO2/Bi2WO6 that effectively enhances visible-light-driven charge separation and redox ability, leading to superior photocatalytic degradation and antibacterial performance. The dual-functionality and high mineralization efficiency demonstrated in this study highlight the potential of the developed photocatalyst for sustainable wastewater treatment and biomedical-related environmental applications. The composite and its constituents were synthesized by hydrothermal approach and characterized in terms of their structural, optical, functional, morphological, photocatalytic and antibacterial properties. The optimized Er-doped ZrO2/Bi2WO6 nanocomposite exhibited excellent photocatalytic performance, achieving 97.02% degradation of the moxifloxacin (MOX) drug under visible light and showed good antibacterial capability against four bacterial strains the order of Escherichia coli > Bacillus fortis > Staphylococcus aureus > Streptococcus canis. Reactive-species scavenging analysis proved that the role of superoxide radicals (center dot O-2(-)) and hydroxyl (center dot OH) plays a significant role in the photocatalytic degradation of MOX. The current study highlights the enhanced photocatalytic and antibacterial activity of the Z-scheme Er doped ZrO2/Bi2WO6 heterostructure due to their design and engineering, with applications in multifunctional industries, particularly environmental cleaning and biomedical research.
Accurate energy production prediction is a critical challenge in modern renewable energy systems due to the inherent variability and complexity of influencing factors. This study proposes an optimized hybrid deep learning framework that integrates Convolutional Neural Networks (CNN) and Recurrent Neural Networks (RNN) enhanced by Genetic Algorithm (GA)-based hyperparameter optimization to improve predictive performance. A publicly available renewable energy dataset comprising 15,000 records and 13 features was utilized, where Z-score normalization was applied to standardize input variables and enhance model convergence. To reduce dimensionality and retain the most informative attributes, Binary Particle Swarm Optimization (BPSO) was employed for feature selection and benchmarked against Binary Grey Wolf Optimization (BGWO), Binary Whale Optimization Algorithm (BWOA), and Binary Butterfly Optimization Algorithm (BBOA), demonstrating superior performance in terms of fitness and error metrics. The dataset was partitioned into training (70%), validation (20%), and testing (10%) subsets to ensure robust generalization. The proposed CNN-RNN model was further optimized using GA and compared with several baseline regression models, including Gradient Boosting Regressor (GBR), Extra Trees Regressor (ETR), Decision Tree Regressor (DTR), K-Nearest Neighbors Regressor (KNNR), and Dummy Regressor (DR). Experimental results indicate that the optimized GA-CNN-RNN model significantly outperformed all competing models, achieving the lowest Mean Squared Error (MSE = 0.0098), Mean Absolute Error (MAE = 0.0791), Median Absolute Error (MedAE = 0.0683), Mean Absolute Percentage Error (MAPE = 0.0090), and the highest coefficient of determination (R2 = 99.85%). Comparative analysis with alternative metaheuristic optimizers, including Artificial Bee Colony (ABC), Firefly Algorithm (FA), and Dragonfly Algorithm (DA), further confirmed the superiority of GA in enhancing model performance.
SARS-CoV-2 main protease (Mpro) remains a clinically validated antiviral target; however, computational hit nomination must distinguish binding-site stability from experimentally confirmed inhibition. This study evaluates a focused positional series of four furan-substituted chromone derivatives (M1-M4) as computationally prioritized Mpro-binding candidates using the high-resolution Mpro co-structure PDB 9C8Q and its co-crystallized ligand A1AU4 as the primary structural reference. The workflow integrates SwissADME and ProTox-III profiling, validated AutoDock Vina docking, 100 ns molecular dynamics simulations of M1-M4 and A1AU4 complexes, endpoint MM/GBSA calculations over 100 equilibrated snapshots from 60-100 ns, per-residue energy decomposition and PCA-based free-energy landscape analysis. All chromone derivatives satisfied Lipinski and Veber criteria and showed low predicted acute toxicity classes. Docking indicated comparable scores for M2 and M3 (-7.0 kcal/mol), but M3 displayed a more specific active-site interaction pattern through conventional hydrogen bonds with His163 and Glu166, pi-sulfur contact with Cys145 and pi-alkyl stabilization near Met165. Comparative dynamics showed stable Mpro backbones in all systems, whereas the co-crystallized ligand remained the most stable ligand reference and M2 displayed the largest ligand RMSD drift. Among the chromone derivatives, M1, M3 and M4 maintained bound-state ensembles suitable for comparative MM/GBSA analysis, with favorable binding supported mainly by van der Waals interactions and moderated by polar desolvation. PCA/FEL analysis further suggested compact bound ensembles without implying experimental antiviral activity. Overall, M3 is proposed as a computationally prioritized chromone scaffold because it combines acceptable drug-likeness, low predicted acute toxicity and specific anchoring within the Mpro catalytic pocket. Biochemical protease assays, antiviral cell assays and toxicity validation are required before any inhibitory or therapeutic claim can be made.
Count data exhibiting overdispersion are common in applied statistical modeling, and the negative binomial regression model is one of the most widely used frameworks for such data. Parameter estimation conventionally relies on maximum likelihood, but this approach deteriorates markedly when explanatory variables are strongly intercorrelated: multicollinearity inflates coefficient variances, compromises estimation precision, increases the mean squared error, and undermines the reliability of statistical inference. To address these limitations, this paper introduces modified Liu estimators tailored to the negative binomial regression setting, offering greater flexibility in handling multicollinearity and yielding more stable results than existing shrinkage approaches. The theoretical properties of the proposed estimators are examined analytically, and formal dominance conditions that establish their superiority over several competing estimators in the existing literature are derived. The practical performance of the proposed estimators is assessed through a structured Monte Carlo simulation experiment and further illustrated via an empirical analysis of CO2 emissions data from Canada. The findings from both the simulation study and the empirical application confirm that the proposed estimators outperform the existing ones, achieving a better balance between bias and variance. In the simulation, the proposed estimators yield lower mean squared error and mean absolute error across all scenarios considered. Similarly, in the empirical application, the proposed estimators produce lower mean squared error and standard errors, along with narrower confidence intervals, compared with the existing estimators.
