Browse Articles

Exact exchange in the Levy–Perdew–Sahni density-functional theory

We show that the Fermi–Amaldi functional constitutes the exact exchange contribution within the orbital-free Levy–Perdew–Sahni (LPS) density-functional scheme, which is based on a Schrödinger-like equation for the square root of the electron density. This result follows from interpreting the LPS scheme in terms of an auxiliary reference system of non-interacting particles occupying a single orbital. Unlike Kohn–Sham exact exchange, the exact LPS exchange is a pure density functional fully compatible with the orbital-free formalism. Self-consistent calculations for atoms and molecules show improved accuracy for light atoms relative to standard Kohn–Sham exchange approximations. These findings provide a basis for developing improved approximations to the remaining terms of the LPS total-energy functional.

A generative adversarial network framework for individualized training load distribution based on physiological response patterns and injury risk prediction

Coherent modeling of double-folded ring polymers and their underlying random tree structure

Topologically constrained genome-like polymers often double-fold into tree-like configurations, which can be modeled on the level of folded (ring) polymers or on the level of the underlying random trees. For both descriptions, we have recently obtained expressions for the configurational entropy in ensembles with controlled branching activity. Here, we demonstrate that they are equivalent up to a contribution originating from the number of distinct wrappings of a single tree. This allows us to develop a coherent framework for freely switching between the two representations. Importantly, the equivalence extends to interacting systems, provided the interactions are treated consistently on the tree and on the ring level. To demonstrate the utility of the scheme, we introduce a generalization of the Amoeba Monte Carlo algorithm capable of generating the required ensembles of trees with fluctuating sizes. While the tree algorithm reproduces results obtained by dynamic simulations of the corresponding ring model, it is O(N) faster for the purpose of sampling static properties and leverages the utility of the ring model for the study of dynamical properties, when used for the preparation of equilibrated starting states.

Mechanostereochemical modulation of polymer mechanical properties

Stereochemistry profoundly affects the physical and mechanical properties of polymers, illustrated by the contrast between elastic natural rubber ( cis -polyisoprene) and the stiffer, less extensible gutta-percha ( trans -isomer). Traditional stereochemistry such as tacticity and cis / trans isomerism primarily governs polymer properties based on fixed structural or conformational factors. Herein, by incorporating the mechanostereochemistry concept into polymers, we demonstrate a stereochemistry paradigm wherein dynamic isomers dictate material properties, thereby defining the unprecedented transient-stereostructure-efficacy mechanism. Specifically, we engineer two mechanically interlocked networks based on [ c 2]daisy chains, where force-triggered intramolecular motion generates mechanostereoisomers with distinct geometric configurations: [ c 2]Daisy chain 1 in MIN- 1 contracts into a fisherman’s knot, whereas [ c 2]daisy chain 2 in MIN- 2 extends into a loop. Due to reduced network elasticity from the loop structure, MIN- 2 exhibits a lower modulus in large-strain shear measurements and less pronounced strain hardening in tensile tests compared to MIN- 1 . Since these mechanostereoisomers are induced by force, material properties show strain-dependent character: both networks perform similarly under small or no strain, but diverge significantly at large strains. Our work expands the conceptual boundaries of polymer stereochemistry and provides insights for designing high-performance materials through stereochemical control.

Determination of worst-case cephalosporin residues in laboratory glassware using a validated low-level RP-HPLC method

Explicitly correlated Gaussian basis approach to periodic systems

Closed-form expressions for all matrix elements required for variational calculation of the electronic structure of periodic solids have been derived using a basis of explicitly correlated Gaussians. Periodic basis functions are constructed by summing shifted correlated Gaussians over all composite lattice translations, where a generalized unfolding theorem reduces the resulting double lattice sum to a single sum through a unified computational framework for overlap, kinetic energy, and Coulomb potential operators. The formalism has been validated through application to an infinite one-dimensional hydrogen chain, where the ground-state energy per atom computed in the thermodynamic limit is shown to agree with finite-chain results extrapolated by many other many-body methods.

Copula-based generation of synthetic multivariate time-series for underground mine gas concentrations

Sensitivity analysis of cycle flux response in nonequilibrium dynamics

The decomposition of edge current into cycle fluxes sheds light on understanding the nonequilibrium steady state structure of stochastic graph models, providing important insights into irreversibility, stability, and dominant functionality of a plethora of nonequilibrium physics. However, the response of the cycle flux to time-dependent perturbations is less well understood. Here, we introduce a theoretical method to analyze the response properties of the cycle flux around the nonequilibrium steady state, rather than the near equilibrium response provided by the well-known linear response theory. We find that both the state and cycle flux responses are determined by the system relaxation spectrum, the signal frequency, as well as the overlap between the signal matrix and the transient states. We further study the relation between response sensitivity and the information gain contained in the stochastic trajectory, providing an information theoretic bound on cycle flux response precision. Furthermore, we use both a quantum Maxwell demon model and a classical chemical reaction network model to illustrate our theory. Our work paves the way toward fully understanding and controlling the dynamical response of complex nonequilibrium stochastic graph models and inferring the hidden time-dependent signals by measuring the observable cycle flux response.

Antimalarial cytoskeletal targeting with broad apicomplexan activity

Malaria is a devastating disease that resulted in an estimated 610,000 deaths in 2024, the majority being children under the age of five. Here, we use KNX-115 to illustrate multistage antiparasitic activity upon targeting the cytoskeletal enzyme Plasmodium falciparum myosin A (PfMyoA). KNX-115 inhibits purified actin-activated ATPase with a potency in the low nanomolar range and >50-fold selectivity against cardiac, skeletal, and smooth muscle myosins. KNX-115 traps PfMyoA in a state that binds weakly to actin. A 2.35 Å resolution structure of KNX-115 bound to PfMyoA reveals critical interactions contributing to its mechanism of action. Importantly, in vitro evolution data reveal that KNX-115 engages PfMyoA as a sole cellular target. Inhibiting PfMyoA blocks the development of the blood and liver stages of laboratory strains of P. falciparum , with no liver cell toxicity, sporozoite cell traversal and motility, and sporozoite development in the mosquito. Inhibiting PfMyoA completely kills parasites after 96 h of treatment. Furthermore, KNX-115 is equally effective at inhibiting a panel of Plasmodium strains resistant to experimental and marketed antimalarials and shows inhibitory activity against P. falciparum circulating isolates from the Brazilian Amazon. Inhibiting PfMyoA with KNX-115 also blocks the blood stage of a laboratory strain of Plasmodium vivax . In line with the evolutionary identity of MyoA among various apicomplexan parasites, KNX-115 also inhibits Cryptosporidium and Eimeria MyoA in vitro and is an effective inhibitor of Cryptosporidium , Toxoplasma, and Eimeria cellular growth, with EC 50 s similar to those found for blood and liver stage Plasmodium .

Ultrasound assessment of the diaphragm and abdominal muscles among elite wrestlers with and without low back pain- a case-control study

Quantifying the relationship between electric field enhancement and plasmon-driven electron transfer

Plasmonic materials interact strongly with light to create localized, out-of-equilibrium environments with intense electromagnetic fields known as hotspots. After forming, hotspots dissipate energy into their surroundings and can transfer energy and charge carriers to nearby molecules, giving plasmonic materials the potential to drive reactions with sunlight. However, the field needs a better mechanistic understanding of plasmon–molecule interactions and how the local plasmon environment, specifically the electromagnetic field enhancement and spatial distribution of hotspots, impacts the reaction yield. In this work, we mapped plasmon-driven charge transfer across ordered plasmonic substrates using diffraction-limited surface-enhanced Raman spectroscopy (SERS) microscopy to understand the relationship between the average local electric field enhancement and charge transfer reaction yield. We tracked the plasmon-induced electron transfer to buckminsterfullerene (C60) and found that areas with the greatest SERS intensity were not the areas with the greatest ensemble-averaged reduction of C60, suggesting that areas with higher electric field enhancement—or “hotter,” more enhancing hotspots—do not improve the charge transfer reaction yield. This work shows that efforts to improve plasmon-driven charge transfer should not merely focus on creating substrates with extremely enhancing regions but also consider how other factors could optimize photoreduction yields.

Resolving weak moments in canted antiferromagnets via time-domain terahertz emission spectroscopy

Canted antiferromagnets host weak net magnetization arising from spin canting while preserving the ultrafast response of antiferromagnetic order. Probing such minute moments—especially in thin films—remains challenging, as conventional magneto-optical and electrical techniques lack sufficient sensitivity. Here, we employ time-domain terahertz emission spectroscopy as a contact-free probe of such weak magnetism. Using epitaxial TmFeO 3 /Pt heterostructures as a model system, the emitted THz signal directly follows the orientation of the canted Fe 3+ moment, allowing real-time tracking of its continuous rotation through the spin reorientation transition. Moreover, the THz hysteresis polarity is reversed compared with conventional ferromagnet/Pt systems, indicating an unconventional interfacial magnetic configuration. Our findings establish THz emission spectroscopy as a powerful tool for resolving ultrafast THz emission response in canted antiferromagnetic thin and ultrathin films.

Cross-validation and cross-cultural invariance of the 10-item athletic mental energy scale

<i>In</i> <i>silico</i> biophysical screening of surface ligands for enhanced nanoparticle mucopermeation

The mucus layer, a physiologically relevant hydrogel found in the human gut, respiratory and genital tracts, and on ocular surfaces, forms a formidable physiological barrier with mucin proteins, creating its hygroscopic mesh that restricts the permeation of nanomedicines. Hydrophilic ligands such as polyethylene glycol, when grafted onto nanoparticles, are known to enhance mucopermeation, although the mechanisms remain unclear. To investigate this, 79 ligands were virtually selected from the ChEMBL and DrugBank databases based on their similarity to triethylene glycol (TEG), an analog of polyethylene glycol, and on Lipinski–Veber oral bioavailability parameters. The ligands were ranked by similarity score (SS), proposed here as a structural bioinformatic parameter for automated ligand-candidate prioritization. A final set of 15 ligands was selected and divided into terciles. The top tercile, which included triethylene glycol, had the highest SS values. These 15 ligands were docked with the equilibrated (unglycosylated) MUC5AC protein structure using the PatchDock and AutoDock Vina programs. For each ligand, the average docking or binding score (depending on the program used) and the average atomic contact energy (mean of the top five atomic contact energies) were calculated. The SS values correlated strongly with the average atomic contact energies and moderately with the average docking and binding scores, as well as with the Lipinski–Veber parameters. These trends were confirmed using 15 additional ligands docked on unglycosylated mucin, with the two top- and two bottom-ranked ligands and TEG further evaluated on glycosylated MUC5AC using PatchDock. Overall, the SS and the average atomic contact energy emerged as effective indicators for efficient and accurate surface-ligand screening.

A unified machine-learning framework for ab initio multiscale modeling of liquids

Understanding and predicting the behavior of liquid matter across length scales—using only the microscopic interactions encoded in the Schrödinger equation—remains a central challenge in the physical sciences. Achieving this goal requires not only an accurate and efficient description of intermolecular forces but also a consistent framework that bridges the micro-, meso-, and macroscales. Here, by combining machine-learned interatomic potentials (MLIPs) with neural classical density functional theory (cDFT), we present such a framework. MLIPs trained on quantum-mechanical energies and forces are used to generate inhomogeneous density profiles, which then serve as the training data for neural cDFT. The resulting ab initio neural cDFT is more computationally efficient than molecular simulations and provides a conceptually transparent route to the thermodynamics of both homogeneous and planar inhomogeneous systems. We demonstrate the approach for both water and carbon dioxide using several exchange–correlation functionals. Beyond accurately reproducing—at the level of the underlying approximate electronic structure—bulk equations of state and liquid–vapor phase diagrams, ab initio neural cDFT predicts, from first principles, how confinement modifies liquid–vapor coexistence in water. It also captures complex behavior in supercritical carbon dioxide such as the Fisher–Widom and Widom lines. While current applications are limited to bulk fluids and planar geometries, this approach establishes a general first-principles route to multiscale modeling of fluids by unifying two independently developed machine-learning paradigms. This work represents an important step toward generalizing cDFT beyond simple empirical potentials to chemically complex systems.

An intelligent synapse fusion network with dynamic adaptive control for multimodal fusion

Electronic structure and metal–metal bonding in early lanthanide dimers La2, Ce2, and Pr2

Understanding whether 4f electrons participate directly in lanthanide–lanthanide bonding remains a long-standing question in f-element chemistry. Lanthanide dimers provide the simplest molecular platform for isolating intrinsic metal–metal bonding interactions without ligand-field or oxidation-state complications. Here, we present a fully relativistic multireference investigation of La2, Ce2, and Pr2 to resolve their ground electronic states and elucidate the evolution of bonding across the early lanthanide series. The ground states under spin–orbit coupling are determined to be 0g, 1g, and 2u for La2, Ce2, and Pr2, respectively, with spectroscopic constants in improved agreement with available experimental data. Molecular orbital analysis reveals a common σ2π4 inner-core framework, while Ce2 exhibits the highest bond order, consistent with partial multiple-bonding character. Importantly, the results demonstrate that 4f orbitals contribute to metal–metal bonding only through cooperative interaction with energetically accessible 5d orbitals; in the absence of such mixing, the 4f electrons remain essentially nonbonding. These findings provide a unified chemical picture for the onset of f-electron participation in lanthanide bonding and clarify the electronic origin of metal–metal interactions in early lanthanide dimers.

Context-aware multimodal AI navigates hidden pathways in five centuries of art evolution

The rise of multimodal generative AI transforms the intersection of technology and art, offering richer insights into large-scale artworks. While significant research has focused on their creative potential, their ability to represent artworks in latent spaces remains underexamined. We use generative AI, specifically Stable Diffusion, to analyze 500 y of Western paintings by extracting two types of latent information with the model: formal aspects (e.g., colors) and contextual aspects (e.g., subjects). Our findings reveal that contextual information exhibits stronger vector alignment and orientation with conventional artistic periods, styles, and individual artists than formal elements. Also, we show how artistic expression aligns with historical shifts using contextual keywords extracted from paintings. Our generative experiment, infusing prospective contexts into historical artworks, validates this vector alignment and orientation by synthesizing artworks consistent with the stylistic patterns of target periods. This study demonstrates how multimodal AI expands traditional formal analysis by integrating temporal, cultural, and historical contexts to quantify the latent structure of cultural knowledge.

LCA analysis of conventional and organic onion production in Poland

Abstract This study presents a comparative analysis of onion production technologies under conventional and organic farming systems, focusing on their environmental impacts. The research covered onion cultivation in southern Poland—a region that is not only one of the country’s leading production areas, but also among the largest in Europe. A “cradle-to-gate” approach was applied, since at this stage of production considerable technological differences exist between the two cultivation systems. The analysis considered the type of field operations, the machines used and their operating time, the amounts of fertilizers, pesticides, fuel, and water consumed. Onion yield was measured, and the results were related both to cultivation area (1 ha) and production volume (1 Mg). To identify the environmental relationships of all inputs and outputs within the LCA scope and to estimate their impacts, the software package SimaPro (version 8.1.0.60) was applied. The results demonstrated greater negative environmental impact of individual operations in conventional onion cultivation compared to organic farming, regardless of the chosen functional unit. Critical processes were identified in terms of environmental burden: harvesting in both systems, and mineral fertilization in the case of conventional cultivation.

VAMP7-dependent mitochondria–lysosome contacts contribute to glial mitochondrial dynamics and dopaminergic neuron survival

Although disrupted mitochondrial dynamics in neurons are closely linked to neurodegenerative diseases, far less is known about how mitochondrial dynamics are regulated in glia or whether glial mitochondrial dysfunction contributes to neurodegeneration. Here, we show that the R-SNARE protein VAMP7 regulates the untethering of mitochondria–lysosome contacts (MLCs) in adult fly glia. Glial-specific knockdown of VAMP7 leads to prolonged MLCs and mitochondrial elongation associated with altered fission/fusion dynamics. These VAMP7-deficient mitochondria exhibit hyperpolarized membrane potential, leading to increased reactive oxygen species production, lipid droplet accumulation, and dopaminergic neurodegeneration. Mechanistically, VAMP7 interacts with the GTPase-activating protein TBC1D15-17 to promote Rab7 GTP hydrolysis. Without VAMP7, TBC1D15-17 remains bound to Rab7 but fails to activate its hydrolysis, resulting in elevated GTP-bound Rab7 and impaired MLCs untethering. Consistently, expression of GTP-locked Rab7 Q67L or GTPase-activating protein-dead TBC1D15-17 ΔGAP phenocopies the mitochondrial defects, while GDP-bound Rab7 T22N or wild-type TBC1D15-17 restores the MLC dynamics. Considering that SNARE proteins mediate membrane fusion, our results demonstrate a role for VAMP7 in glial mitochondrial dynamics via organelle contacts, impacting neuron survival in a non-cell-autonomous manner.