Browse Articles

Anomalous expansion of interatomic distance in liquid Al–Zn alloy during cooling

The structural and physical/chemical properties of metallic materials are closely linked to the composition and configuration of their molten state. In this study, the evolution of the local structure in Al–Zn alloys with varying compositions during the quenching process was investigated using on-the-fly machine learning force field (MLFF) simulations based on ab initio molecular dynamics. The results indicate that the first coordination shell of the Al–Zn alloy melt undergoes an anomalous expansion within a specific temperature range, which deviates from the previously reported linear negative expansion, such as metallic Al, Zn, and Sn. The temperature interval of the anomalous expansion decreases with increasing concentration of Zn. Local structural changes, including the abnormal increase in coordination number and the emergence of a shoulder in the second peak of the bond angle distribution function, further confirmed the presence of the anomalous expansion. In addition, the slope of the energy–temperature relationship and the activation energy of diffusion change upon temperature decrease, which suggests that this phenomenon is associated with atomic diffusion driven by thermal excitation. The anomalous expansion is also confirmed in a larger system (over 104 atoms) by MLFF simulations with first-principles accuracy. We found that an increase in cluster radius induces a decrease in pressure due to the interfacial energy of the cluster at ∼1200–1100 K for Al–38Zn melts, which in turn leads to a looser arrangement of the atoms. This study provides valuable insights into the mechanisms governing atomic structure evolution and liquid–liquid transitions.

Correction: Horse handlers’ knowledge, attitudes, and perceptions of African horse sickness in South-West, Nigeria

Circumventing problems introduced by matrix asymmetry in collocation calculations of vibrational spectra by exploiting near symmetry

Collocation is an enticing alternative to variational methods for solving the vibrational Schrödinger equation. It makes it possible to use a general potential without requiring integrals and quadrature. An important disadvantage of collocation is the need to work with nonsymmetric matrices. Eigenvalues of a large matrix are best computed with an iterative method, but iterative eigensolvers are much more efficient for symmetric matrices. Heretofore, it has been costly to use collocation when the basis set and Hamiltonian matrix are large. We demonstrate that it is possible to systematically make the collocation matrix whose eigenvalues one must compute more and more symmetric and propose an efficient iterative eigensolver for a nearly symmetric matrix. Little is known about exploiting near symmetry. We use a combination of filter diagonalization and an iterative linear solver powered by a three-term recursion relation. We test the ideas with a 6-D Hamiltonian and show that accurate energies are obtained despite the asymmetry.

Robust zero-watermarking for color images using hybrid deep learning models and encryption

Abstract Reliable zero-watermarking is a distortion-free approach to copyright protection, which has been a primary focus of digital watermarking research. Traditional zero-watermarking techniques often struggle to maintain resilience against geometric and signal processing attacks while ensuring high security and imperceptibility. Many existing methods fail to extract stable and distinguishable features, making them vulnerable to image distortions such as compression, filtering, and geometric transformations. This paper presents a robust zero-watermarking technique for color images, combining Local Binary Patterns (LBP) with deep features extracted from the CONV5-4 layer of the VGG19 neural network to overcome these limitations. Frequent domain transformations, utilizing the Discrete Wavelet Transform (DWT) and Discrete Cosine Transform (DCT), enhance feature representation and improve resilience. Furthermore, a chaotic encryption scheme based on the Lorenz system and the Logistic map is used to scramble the feature matrix and watermark, thereby ensuring increased security. The zero watermark is generated through an XOR operation, facilitating imperceptible and secure ownership verification. Experimental results show that the proposed method is highly resilient to various attacks, including scaling, noise, filtering, compression, and rotation. The extracted watermark maintains a low Bit Error Rate (BER) and a high Normalized Cross-Correlation (NCC). At the same time, the Peak Signal-to-Noise Ratio (PSNR) of attacked images remains optimal. Specifically, the BER values of the extracted watermarks were below 0.0022, and the NCC values were above 0.9959. In contrast, the average PSNR values of the attacked images reached 34.0692 dB, demonstrating the method’s superior robustness and visual quality. Compared to existing zero-watermarking algorithms, the proposed method shows superior robustness and security, making it highly effective for multimedia copyright protection.

Transport mechanism of fluorosulfonylamide-based molten alkali metal salts—Intermediate temperature ionic liquids

Molecular dynamics (MD) simulations in this study elucidated the transport mechanism of a series of intermediate temperature ionic liquids: molten MFSA, MFTA, and MTFSA, where M = (Li, Na, K, Rb, and Cs), FSA = bis(fluorosulfonyl)amide, FTA = fluorosulfonyl(trifluoromethylsulfonyl)amide, and TFSA = bis(trifluoromethylsulfonyl)amide. The following two peculiarities had been experimentally observed: (i) the electrical conductivity, σ, of Li-systems is extremely lower than that of the other alkali metal counterparts and (ii) the Nernst–Einstein conductivity, σNE, derived from the self-diffusion coefficients, D+ and D−, of LiFSA and LiFTA is lower than the real conductivity, σ > σNE, which is usually the other way around. Hypothetical MD simulations made by increasing the size or decreasing the valence of Li+ revealed that the strong interaction between neighboring cation and anion caused by the high surface charge density on Li+ is responsible for both (i) and (ii). Theoretical consequences derived from the momentum conservation and the separation of σ into its components in terms of velocity correlation coefficients proved that, in addition to these features, the significant mass difference between a cation and anion for the Li-systems leads to (iii) the almost exclusive contribution of Li+ to σ and (iv) a positive contribution of the Li+–Li+ cross correlation, which is negative for other systems. Hypothetical simulations at high temperatures, at which the anions actually decompose, suggested that features (i), (ii), and (iv) stem from the “intermediate” temperature range at which these salts are fluid.

Tolerant integrated reciprocity sustains cooperation in a noisy environment

Isomerization-assisted proton transfers in MeOH-(H2O)2H+

We carried out Path Integral Molecular Dynamics simulations that describe the microscopic properties of two isomerization processes taking place in the MeOH(H2O)2H+ trimer, at T = 50 K. In particular, we focused attention on the free energies associated with the exchange of the connective/dangling characteristics of a pair of protons located at key subunits of the trimer. In one of the processes, the isomerization produces a modification in the topology of the cluster’s connectivity pattern, from branched-like to chain-like motifs. In contrast, along the other transformation, reactant and product states are both chain-like and equivalent. Changes in associated free energies were computed following reversible paths described in terms of order parameters involving angular degrees of freedom. As a common feature, along both isomerizations, we registered concomitant migrations of the excess proton. The strongly confining, single-well characteristics of the potential energy surface along the asymmetric stretch coordinate promote compact, ∼0.35 Å long structures for the isomorphic polymer associated with the itinerant proton, which persist along the complete isomerization path. These observations suggest the absence of tunneling contributions to the resulting mechanisms that control the proton transfer process. Estimates for the corresponding isomerization rates are also computed.

Exome analysis links kidney malformations to developmental disorders and reveals causal genes

The dissemination potential of Microsporidia MB in Anopheles arabiensis mosquitoes is modulated by temperature

Abstract Microsporidia MB , a vertically transmitted endosymbiont of Anopheles mosquitoes, shows strong potential as a malaria control agent due to its ability to inhibit Plasmodium development within the mosquito host. To support its deployment in malaria transmission reduction strategies, it is critical to understand how environmental factors, particularly temperature, influence its infection dynamics. In this study, we investigated the impact of four temperature regimes (22 °C, 27 °C, 32 °C, and 37 °C) on Microsporidia MB prevalence and infection intensity by rearing mosquito larvae under controlled laboratory conditions. Our results demonstrate that elevated temperatures, especially 32 °C, significantly enhance both larval growth and Microsporidia MB infection rates. Population growth modeling further indicates that at 32 °C, an infected mosquito population can reach 1000 offspring within 15–35 days, representing a 4.7-, 1.3-, and 1.7-fold increase in dissemination potential compared to 22 °C, 27 °C, and 37 °C, respectively. Although mortality at 32 °C was approximately 20% higher than at 27 °C, this temperature emerged as the most favorable for mass-rearing Microsporidia MB -infected larvae. These findings provide the first insights into temperature-mediated dynamics of Microsporidia MB and support its potential for scalable implementation in malaria-endemic regions.

On the influence of bending energy on the assembly of spherical viral capsids

The protective shell, or capsid, of many spherical viruses is formed via a self-assembly process whose underlying physical principles have not yet been fully elucidated. In this article, we analyze the role of elastic bending energy in the in vitro self-assembly of a spherical capsid in the limit where such energetic contribution dominates over compression stress. The model predicts that the capsid closes prematurely, and its final size is completely determined by a dimensionless constant fr, which is the ratio of the bending modulus to the line tension of the edge. In addition, we compute the critical size, the nucleation barrier, and the assembly rate of capsids and compare our results with those previously obtained by the original classical nucleation theory of viral capsids, where the elastic energy was neglected. Our model suggests that the competition between line tension and bending energy accelerates the rate of capsid nuclei production and causes capsids to close at suboptimal sizes, suggesting that capsomers have optimal bending angles that differ from the values measured in native viruses.

Joint, multifaceted genomic analysis enables diagnosis of diverse, ultra-rare monogenic presentations

Abstract Genomics for rare disease diagnosis has advanced at a rapid pace due to our ability to perform in-depth analyses on individual patients with ultra-rare diseases. The increasing sizes of ultra-rare disease cohorts internationally newly enables cohort-wide analyses for new discoveries, but well-calibrated statistical genetics approaches for jointly analyzing these patients are still under development. The Undiagnosed Diseases Network (UDN) brings multiple clinical, research and experimental centers under the same umbrella across the United States to facilitate and scale case-based diagnostic analyses. Here, we present the first joint analysis of whole genome sequencing data of UDN patients across the network. We introduce new, well-calibrated statistical methods for prioritizing disease genes with de novo recurrence and compound heterozygosity. We also detect pathways enriched with candidate and known diagnostic genes. Our computational analysis, coupled with a systematic clinical review, recapitulated known diagnoses and revealed new disease associations. We further release a software package, RaMeDiES, enabling automated cross-analysis of deidentified sequenced cohorts for new diagnostic and research discoveries. Gene-level findings and variant-level information across the cohort are available in a public-facing browser ( https://dbmi-bgm.github.io/udn-browser/ ). These results show that case-level diagnostic efforts should be supplemented by a joint genomic analysis across cohorts.

Exploring the clinical value of concept-based AI explanations in gastrointestinal disease detection

Abstract Complex artificial intelligence models, like deep neural networks, have shown exceptional capabilities to detect early-stage polyps and tumors in the gastrointestinal tract. These technologies are already beginning to assist gastroenterologists in the endoscopy suite. To understand how these complex models work and their limitations, model explanations can be useful. Moreover, medical doctors specialized in gastroenterology can provide valuable feedback on the model explanations. This study explores three different explainable artificial intelligence methods for explaining a deep neural network detecting gastrointestinal abnormalities. The model explanations are presented to gastroenterologists. Furthermore, the clinical applicability of the explanation methods from the healthcare personnel’s perspective is discussed. Our findings indicate that the explanation methods are not meeting the requirements for clinical use, but that they can provide valuable information to researchers and model developers. Higher quality datasets and careful considerations regarding how the explanations are presented might lead to solutions that are more welcome in the clinic.

Non-equilibrium origin of cavity-induced resonant modifications of chemical reactivities

In this work, we investigate the influence of light–matter coupling on reaction dynamics and equilibrium properties of a single molecule inside an optical cavity. The reactive molecule is modeled using a triple-well potential, allowing two competing reaction pathways that yield distinct products. Dynamical and equilibrium simulations are performed using the numerically exact hierarchical equations of motion approach in real- and imaginary-time formulations, respectively, both implemented with tree tensor network decomposition schemes. We consider two illustrative cases: one dominated by slow kinetics and another by ultrafast processes. Our results demonstrate that the rates of ground-state reaction pathways can be selectively enhanced when the cavity frequency is tuned into resonance with a vibrational transition directly leading to the formation of the corresponding product, even when that transition is spectroscopically dark. However, tuning cavity frequency to match an absorption-dominant transition shared across both reaction pathways does not necessarily result in pronounced rate enhancements and selectivity. Together with an additional analysis using an asymmetric double-well model, we highlight the greater complexity of underlying factors governing chemical reactivity, which extend beyond considerations of transition dipole strengths and thermal population distributions that shape linear spectroscopy. Furthermore, we found that in all scenarios, the equilibrium populations remain unchanged when the molecule is moved into the cavity, regardless of the cavity frequency. Thus, our proof-of-concept study confirms at a fully quantum-mechanical level that cavity-induced modifications of chemical reactivities in resonant conditions arise from dynamical and non-equilibrium interactions between the cavity mode and molecular vibrations, rather than from the significant changes in equilibrium properties.

Crosstalk between inovirus core gene and accessory toxin-antitoxin system mediates polylysogeny

Advanced skin cancer prediction with medical image data using MobileNetV2 deep learning and optimized techniques

Vibronic spectrum of pyrazine: New insights from multi-state-multi-mode simulations parameterized with equation-of-motion coupled-cluster methods

This study reports simulations of the lowest band in the electronic absorption spectrum of pyrazine carried out using a multi-state-multimode vibronic Hamiltonian parameterized using equation-of-motion coupled-cluster methods. The simulations explain the main spectral features and show how peaks of vibronic nature appear. The most complete vibronic model includes four electronic states and six vibrational modes. The simulations reveal that non-adiabatic coupling with bright states located as high as 3 eV above the studied state can lead to discernible features in the absorption spectrum. This study demonstrates the power of fully ab initio treatments of electronic and vibrational structure and their utility in understanding the mechanisms leading to complex molecular spectra.

PABPC1 SUMOylation enhances cell survival by promoting mitophagy through stabilizing U-rich mRNAs within stress granules

Explainable illicit drug abuse prediction using hematological differences

Mitigating error cancellation in density functional approximations via machine learning correction

The integration of machine learning (ML) with density functional theory has emerged as a promising strategy to enhance the accuracy of density functional methods. While practical implementations of density functional approximations (DFAs) often exploit error cancellation between chemical species to achieve high accuracy in thermochemical and kinetic energy predictions, this approach is inherently system-dependent, which severely limits the transferability of DFAs. To address this challenge, we developed a novel ML-based correction to the widely used B3LYP functional, directly targeting its deviations from the exact exchange-correlation functional. By utilizing highly accurate absolute energies as exclusive reference data, our approach eliminates the reliance on error cancellation. To optimize the ML model, we attribute errors to real-space pointwise contributions and design a double-cycle protocol that incorporates self-consistent field calculations into the training workflow. Numerical tests demonstrate that the ML model, trained solely on absolute energies, improves the accuracy of calculated relative energies, demonstrating that robust DFAs can be constructed without resorting to error cancellation. Comprehensive benchmarks further show that our ML-corrected B3LYP functional significantly outperforms the original B3LYP across diverse thermochemical and kinetic energy calculations, offering a versatile and superior alternative for practical applications.

Author Correction: An extensive disulfide bond network prevents tail contraction in Agrobacterium tumefaciens phage Milano