Atomic Interactions Research Articles

MGATAF: multi-channel graph attention network with adaptive fusion for cancer-drug response prediction

BackgroundDrug response prediction is critical in precision medicine to determine the most effective and safe treatments for individual patients. Traditional prediction methods relying on demographic and genetic data often fall short in accuracy and robustness. Recent graph-based models, while promising, frequently neglect the critical role of atomic interactions and fail to integrate drug fingerprints with SMILES for comprehensive molecular graph construction.ResultsWe introduce multimodal multi-channel graph attention network with adaptive fusion (MGATAF), a framework designed to enhance drug response predictions by capturing both local and global interactions among graph nodes. MGATAF improves drug representation by integrating SMILES and fingerprints, resulting in more precise predictions of drug effects. The methodology involves constructing multimodal molecular graphs, employing multi-channel graph attention networks to capture diverse interactions, and using adaptive fusion to integrate these interactions at multiple abstraction levels. Empirical results demonstrate MGATAF’s superior performance compared to traditional and other graph-based techniques. For example, on the GDSC dataset, MGATAF achieved a 5.12% improvement in the Pearson correlation coefficient (PCC), reaching 0.9312 with an RMSE of 0.0225. Similarly, in new cell-line tests, MGATAF outperformed baselines with a PCC of 0.8536 and an RMSE of 0.0321 on the GDSC dataset, and a PCC of 0.7364 with an RMSE of 0.0531 on the CCLE dataset.ConclusionsMGATAF significantly advances drug response prediction by effectively integrating multiple molecular data types and capturing complex interactions. This framework enhances prediction accuracy and offers a robust tool for personalized medicine, potentially leading to more effective and safer treatments for patients. Future research can expand on this work by exploring additional data modalities and refining the adaptive fusion mechanisms.

Unbinding of Alpha Chain of Hemoglobin in Sickle and Normal Structures

Abstract Sickle cell disease, a genetic disorder, is caused by a mutation of glutamic acid into valine in the β chain of hemoglobin at the sixth residue, resulting in structural change of the entire hemoglobin molecule into a sickle shape. We investigated the atomic level interaction between the α chain (chain A) and the remaining three chains to identify the structural modification in sickle hemoglobin using the molecular dynamics simulations. H-bond, solvent accessible surface area (SASA), hydrophobic interactions, salt bridges of sickle and normal hemoglobin have been estimated. The estimated parameters from sickle hemoglobin are compared to normal hemoglobin structure. Steered molecular dynamics (SMD) is utilized to estimate the force required in breaking hydrogen bonds in given chains. The SMD simulations at different pulling velocities show that the decoupling forces depend on values of pulling force. This relation is linear, 6780 pN to 12345 pN with pulling velocities of 0.00020 nm/ps to 0.00040 nm/ps in sickle hemoglobin. Much higher force of 8738 pN to 16557 pN in normal is required in normal hemoglobin with same spring constants values from k = 500 to 1100 kcal mol-1 nm-2 and same pulling velocities.

EOSnet: Embedded Overlap Structures for Graph Neural Networks in Predicting Material Properties.

Graph Neural Networks (GNNs) have emerged as powerful tools for predicting material properties, yet they often struggle to capture many-body interactions and require extensive manual feature engineering. Here, we present EOSnet (Embedded Overlap Structures for Graph Neural Networks), a novel approach that addresses these limitations by incorporating Gaussian Overlap Matrix (GOM) fingerprints as node features within the GNN architecture. Unlike models that rely on explicit angular terms or human-engineered features, EOSnet efficiently encodes many-body interactions through orbital overlap matrices, providing a rotationally invariant and transferable representation of atomic environments. The model demonstrates superior performance across various prediction tasks of materials' properties, achieving particularly notable results in properties sensitive to many-body interactions. For band gap prediction, EOSnet achieves a mean absolute error of 0.163 eV, surpassing previous state-of-the-art models. The model also excels in predicting mechanical properties and classifying materials, with 97.7% accuracy in metal/nonmetal classification. These results demonstrate that embedding GOM fingerprints into node features enhances the ability of GNNs to capture complex atomic interactions, making EOSnet a powerful tool for materials' discovery and property prediction.

Programmable Electromagnetic Wave Absorption via Tailored Metal Single Atom-Support Interactions.

Metal single atoms (SA)-support interactions inherently exhibit significant electrochemical activity, demonstrating potential in energy catalysis. However, leveraging these interactions to modulate electronic properties and extend application fields is a formidable challenge, demanding in-depth understanding and quantitative control of atomic-scale interactions. Herein, in situ, off-axis electron holography technique is utilized to directly visualize the interactions between SAs and the graphene surface. These interactions facilitate the formation of dispersed nanoscale regions with high charge density and are highly sensitive to external electromagnetic (EM) fields, resulting in controllable dynamic relaxation processes for charge accumulation and restoration. This leads to customized dielectric relaxation, which is difficult to achieve with current band engineering methods. Moreover, these electronic behaviors are insensitive to elevated temperatures, having characteristics distinct from those of typical metallic or semiconducting materials. Based on these results, programmable EM wave absorption properties are achieved by developing a library of SA-graphene materials and precisely controlling SA-support interactions to tailor their responses to EM waves in terms of frequency and intensity. This advancement addresses the customized anti-EM interference requirements of electronic components, greatly enhancing the development of integrated circuits and micro-nano chips. Future efforts will concentrate on manipulating atomic interactions in SA-support, potentially revolutionizing nanoelectronics and optoelectronics.

Strong signature of right-handed circularly polarized photoionization close to the cyclotron line in the atmosphere of magnetic white dwarfs

Magnetic fields break the symmetry of the interaction of atoms with photons with different polarizations, yielding chirality and anisotropy properties. The dependence of the absorption spectrum on the polarization, a phenomenon known as dichroism, is present in the atmosphere of magnetic white dwarfs. Its evaluation for processes in the continuum spectrum has been elusive so far due to the absence of appropriate ionization equilibrium models and incomplete data on photoionization cross sections. We combined rigorous solutions to the equilibrium of atomic populations with approximate cross sections to calculate the absolute opacity due to photoionization in a magnetized hydrogen gas. We predict a strong right-handed circularly polarized absorption (χ+) formed blueward of the cyclotron resonance for fields from about 14 to several hundred megagauss. In energies lower than the cyclotron fundamental, this absorption shows a deep trough with respect to linear and left-handed circular polarizations that steepens with the field strength. The jump in χ+ is due to the confluence of a large number of photoionization continua produced by right-handed circularly polarized transitions from atomic states with a nonnegative magnetic quantum number toward different Landau levels.

Unraveling the interaction of Ta atoms with Pt(111)

Unraveling the interaction of Ta atoms with Pt(111)

Unravelling the in vivo biotoxicity of green-biofabricated Graphene Oxide-Microplastic hybrid mediated by proximal intrinsic atomic interaction

Graphene Oxide (GO) nano-sheets have emerged as a potent nanomaterial for a range of applications like antibacterial, antibiofilm. Besides, Microplastic are emerging as a chronic pollutant originated from the aggrandized...

Additive Manufacturing with Cellulose‐Based Composites: Materials, Modeling, and Applications

AbstractRecent advances in large‐scale additive manufacturing (AM) with polymer‐based composites have enabled efficient production of high‐performance materials. Cellulose nanomaterials (CNMs) have emerged as bio‐based feedstocks due to their exceptional strength and sustainability. However, challenges such as hornification and poor dispersion in polymer matrices still limit large‐scale CNM–polymer composite manufacturing, requiring novel strategies. This review outlines an approach starting with atomic‐level simulations to link molecular composition to key parameters like bulk density, viscosity, and modulus. These simulations provide data for finite element analysis (FEA), which informs large‐scale experiments and reduces the need for extensive trials. The strategy explores how atomic interactions impact the morphology, adhesion, and mechanical properties of CNM‐based composites in AM processes. The review also discusses current developments in AM, along with predictions of mechanical and thermal properties for structural applications, packaging, flexible electronics, and hydrogel scaffolds. By integrating experimental findings with molecular dynamics (MD) simulations and finite element modeling (FEM), valuable insights for material design, process optimization, and performance enhancement in CNM‐based AM are provided to address ongoing challenges.

The importance of sampling the dynamical modes: Reevaluating benchmarks for invariant and equivariant features of machine learning potentials for simulation of free energy landscapes.

Machine learning interatomic potentials (MLIPs) are rapidly gaining interest for molecular modeling, as they provide a balance between quantum-mechanical level descriptions of atomic interactions and reasonable computational efficiency. However, questions remain regarding the stability of simulations using these potentials, as well as the extent to which the learned potential energy function can be extrapolated safely. Past studies have encountered challenges when MLIPs are applied to classical benchmark systems. In this work, we show that some of these challenges are related to the characteristics of the training datasets, particularly the inefficient exploration of the dynamical modes and the inclusion of rigid constraints. We demonstrate that long stability in simulations with MLIPs can be achieved by generating unconstrained datasets using unbiased classical simulations, provided that the important dynamical modes are correctly sampled. In addition, we emphasize that in order to achieve precise energy predictions, it is important to resort to enhanced sampling techniques for dataset generation, and we demonstrate that safe extrapolation of MLIPs depends on judicious choices related to the system's underlying free energy landscape and the symmetry features embedded within the machine learning models.

Enhancing CO2 Adsorption on MgO: Insights into Dopant Selection and Mechanistic Pathways

Inspired by our recent success in designing CO2-phobic and CO2-philic domains on nano-MgO for effective CO2 adsorption, our ongoing efforts focus on incorporating dopants into pristine MgO to further enhance its CO2 adsorption capabilities. However, a clear set of guidelines for dopant selection and a holistic understanding of the underlying mechanisms is still lacking. In our investigation, we combined first-principles calculations with experimental approaches to explore the crystal and electronic structural changes in MgO doped with high-valence elements (Al, C, Si, and Ti) and their interactions with CO2. Our findings unveiled two distinct mechanisms for CO2 capture: Ti-driven catalytic CO2 decomposition and CO2 polarization induced by Al, C, and Si. Ti doping induced outward Ti atom displacement and structural distortion, facilitating CO2 dissociation, whereas C doping substantially bolstered the electron donation capacity and CO2 adsorption energy. Pristine and C-doped MgO engaged CO2 through surface O atoms, while Al-, Si-, and Ti-doped MgO predominantly relied on dopant–O atom interactions. Our comprehensive research, integrating computational modeling and experimental work supported by scanning electron microscopy and thermal gravimetric analysis, confirmed the superior CO2 adsorption capabilities of C-doped MgO. This yielded profound insights into the mechanisms and principles that govern dopant selection and design.

A new class of penicillin-binding protein inhibitors to address drug-resistant Neisseria gonorrhoeae.

β-Lactams are the most widely used antibiotics for the treatment of bacterial infections because of their proven track record of safety and efficacy. However, susceptibility to β-lactam antibiotics is continually eroded by resistance mechanisms. Emerging multidrug-resistant (MDR) Neisseria gonorrhoeae strains possessing altered penA alleles (encoding PBP2) pose a global health emergency as they threaten the utility of ceftriaxone, the last remaining outpatient antibiotic. Here we disclose a novel benzoxaborinine-based penicillin-binding protein inhibitor series (boro-PBPi) that is envisioned to address penA-mediated resistance while offering protection against evolution and expansion of β-lactamases. Optimization of boro-PBPi led to the identification of compound 21 (VNRX-14079) that exhibits potent antibacterial activity against MDR N. gonorrhoeae achieved by high affinity binding to the PBP2 target. Boro-PBPi/PBP2 complex structures confirmed covalent interaction of the boron atom with Ser310 and the importance of the β3-β4 loop for improved affinity. 21 elicits bactericidal activity, a low frequency of resistance, a good safety profile, suitable pharmacokinetic properties, and in vivo efficacy in a murine infection model against ceftriaxone-resistant N. gonorrhoeae. 21 is a promising anti-gonorrhea agent poised for further advancement.

A high-dimensional neural network potential for Co3O4

The Co3O4spinel is an important material in oxidation catalysis. Its properties under catalytic conditions, i.e. at finite temperatures, can be studied by molecular dynamics simulations, which critically depend on an accurate description of the atomic interactions. Due to the high complexity of Co3O4, which is related to the presence of multiple oxidation states of the cobalt ions, to dateab initiomethods have been essentially the only way to reliably capture the underlying potential energy surface, while more efficient atomistic potentials are very challenging to construct. Consequently, the accessible length and time scales of computer simulations of systems containing Co3O4are still severely limited. Rapid advances in the development of modern machine learning potentials (MLPs) trained on electronic structure data now make it possible to bridge this gap. In this work, we employ a high-dimensional neural network potential (HDNNP) to construct a MLP for bulk Co3O4spinel based on density functional theory calculations. After a careful validation of the potential, we compute various structural, vibrational, and dynamical properties of the Co3O4spinel with a particular focus on its temperature-dependent behavior, including the thermal expansion coefficient.

Continuous cavity QED with an atomic beam

Atoms coupled to cavities provide an exciting playground for the study of fundamental interactions of atoms mediated through a common channel. Many of the applications of cavity QED and cold-atom experiments more broadly, suffer from limitations caused by the transient nature of an atomic loading cycle. The development of continuous operation schemes is necessary to push these systems to the next level of performance. Here we present a machine designed to produce a continuous flux of collimated atoms that traverse an optical cavity. The atom-light interaction is enhanced by a fast-decaying cavity optimal for studying phenomena where atomic properties dominate. We demonstrate the transition to a collective strong-coupling regime heralded by a normal-mode splitting. We observe a second phase with a binary normal-mode splitting born from an offset in the mean velocity of the atoms. Inverting the atomic ensemble in the collective strong-coupling regime, we measure continuous optical gain. This work sets the stage for studying threshold conditions for continuous collective phenomena, such as continuous superradiant lasing. Published by the American Physical Society 2024

Evaluations of the Perturbation Resistance of the Deep-Learning-Based Ligand Conformation Optimization Algorithm.

In recent years, the deep learning (DL) technique has rapidly developed and shown great success in scoring the protein-ligand binding affinities. The protein-ligand conformation optimization based on DL-derived scoring functions holds broad application prospects, for instance, drug design and enzyme engineering. In this study, we evaluated the robustness of a DL-based ligand conformation optimization protocol (DeepRMSD+Vina) for optimizing structures with input perturbations by examining the predicted ligand binding poses and scoring. Our results clearly indicated that compared to traditional optimization algorithms (such as Prime MM-GBSA and Vina optimization), DeepRMSD+Vina exhibits higher performance when treating diverse protein-ligand cases. The DeepRMSD+Vina is robust and can always generate the correct binding structures even if perturbations (up to 3 Å) are introduced to the input structure. The success rate is 62% for perturbation with a RMSD within 2-3 Å. However, the success rate dramatically drops to 11% for large perturbations, with RMSD extending to 3-4 Å. Furthermore, compared to the widely used optimization protocol of AutoDock Vina, the DL-generated conformation shows a balanced performance for all of the systems under examination. Overall, the DL-based DeepRMSD+Vina is unarguably more reliable than the traditional methods, which is attributed to the physically inspired design of the neural networks in DeepRMSD+Vina where the distance-transformed features describing the atomic interactions between the protein and the ligand have been explicitly considered and modeled. The outstanding robustness of the DL-based ligand conformational optimization algorithm further validates its superiority in the field of conformational optimization.

Organosilicon Polymeric Acetylene Derivatives: X-Ray Spectral Study and Quantum-Chemical Calculations

The atomic and electronic structure of two organosilicon polymers [–Ph2Si–(C≡C)2–]n (P1) and [–Ph2Si–(C≡C–C≡C)2–]m (P2) (where Ph — is phenyl group) of acetylene and diacetylene types by methods of density functional theory and X-ray emission spectroscopy. The interpretation of X-ray emission SiKβ1-spectra of these polymers was carried out based on an analysis of the distribution of partial electronic states obtained from quantum chemical calculations. Quantitative characteristics of the parameters of the chemical interaction of atoms, such as populations, natural charges, and electronic configurations in the studied polymers, were obtained based on the analysis of hybrid natural bond orbitals. The obtained values of the natural bond orbitals polarization coefficients indicate that the electron density is localized predominantly on carbon atoms. The electronic configurations for carbon atoms in different fragments differ significantly. For C atoms of ethynyl (diethynyl) fragments, they are close to linear σ-bond with sp1.03 (P1) and sp0.95 (P2) configurations, while for C atoms of phenyl fragments it is sp2.42, intermediate between sp2 and sp3 configurations. The natural charges on Si in both polymers are almost the same: +1.58e, +1.59e, while the natural charges on the carbon atoms of the diethynyl group decrease in comparison with the charge on the carbon atom of the ethynyl group from –0.42e to –0.36e.

Electromagnetic Wave Propagation in Photonic Nanoplates

In this study, an investigation into the behaviour of electromagnetic wave propagation in a nanoplate, utilizing Eringen’s model, which is rooted in nonlocal theory, is conducted. This model accounts for the interaction of atoms that are at a significant distance from each other. The objective of this study is to derive solutions that incorporate both local and nonlocal calculations. The focus of this study revolves around the material property parameter values of the nanoplate and the impact of nano coefficients (nonlocal parameters) on the electromagnetic wave propagation behaviour. Additionally, the frequency and wave amplitude values of the electromagnetic wave propagation are determined. By implementing Eringen’s model, a comprehensive understanding of how electromagnetic waves traverse a nanoplate, considering the influence of nano coefficients on the electromagnetic wave propagation behaviour, and identifying the frequency and amplitude of the electromagnetic waves as they propagate through the nanoplate is explored. The solutions encompass both local and nonlocal calculations are aimed to derive. By using this model, a solution that accounts for both local and nonlocal calculations and examines how electromagnetic waves travel through a nanoplate, the effect of nano coefficients on electromagnetic wave propagation behaviour, and the frequency and amplitude of the electromagnetic waves as they propagate through the nanoplate is obtained.

Instabilities along the Axis of the Heliospheric Jets

Abstract The heliosphere is influenced by the interaction of neutral hydrogen atoms from the interstellar medium with protons in the heliosheath (HS). The confinement by the solar magnetic field of the HS plasma forms distinct north and south heliospheric jets. Previous global MHD simulations of the heliosphere reveal that instabilities can develop in the heliospheric jets and make them become unstable, giving rise to large-scale turbulence. In this study, we show that there is a low-speed region of solar wind in the HS that provides a precondition for instabilities to develop. The low-speed region is formed due to charge exchange and magnetic tension. This region allows sufficient time for the instabilities to develop and induces a shear flow inside the HS leading to the development of a Kelvin–Helmholtz (K-H) instability. The estimated growth timescale of the K-H instability based on the simulation results is around 5–7 yr. Understanding the development of the instabilities in the HS is crucial for comprehending the dynamic processes within the HS and the structure of the heliosphere.

Molecular Insights into Binary Ionic Melts of Protic Ionic Liquid 1,2,4-Triazolium Methanesulfonate and Methanesulfonic Acid Electrolytes.

Binary ionic melts formed by a protic ionic liquid (PIL) 1,2,4-triazolium methanesulfonate ([TAZ][MS]) dissolved in methanesulfonic acid are studied as non-stoichiometric electrolytes. The composition-driven structure-property relationship of methanesulfonic acid is explored at varying molar fraction ratios from 0/100 to 10/90, 20/80, and 30/70 by the addition of 1,2,4-triazolium methanesulfonate [TAZ][MS] IL. To unveil molecular characteristics of these mixtures of [TAZ][MS] PIL and CH3SO3H, we performed classical molecular dynamics simulations at varying temperatures from 293 to 303, 363, and 423 K. Ion-pair interactions and ion-solvent interactions are captured through various atom-atom radial distribution functions, and hydrogen bonding is presented through N-H hydrogen atom interactions with the [MS]- anion or CH3SO3H oxygen atoms. Apart from the temperature, the PIL ratio itself plays a critical role in hydrogen bonding interactions, and spatial density distribution maps show strengthening of ion-pair interactions with an increase in PIL concentration. The mobility of ions and CH3SO3H is also explored through mean square displacement plots and by calculating the diffusion coefficient. The diffusion coefficient is used to calculate Nernst-Einstein conductivity, which shows an excellent match with experimental conductivity at 363 and 423 K.

Dynamic behavior and stability analysis of fluid-conveying nano-spiral shells: Modeling, simulation, and parametric study

This paper presents a study on a fluid-conveying nano-spiral shell’s free vibration and stability analysis. The nano-spiral shell with Archimedean cross-section shape is modeled according to the first-order shear deformation (FSDT) and modified coupled stress (MCST) theories. Van der Waals force is modeled based on the Leonard Jones potential for atomic interactions. The fluid-structure interaction equation is derived based on the assumption of slip boundary conditions for the fluid. The Rayleigh-Ritz method is used to obtain the natural frequencies of a macro-size spiral shell to be validated with the finite element method (FEM). The parametric vibration analysis is performed for a nano-spiral shell conveying fluid. The results show that the natural frequencies decrease as the distance between the layers of the nano-spiral shell increases. Additionally, an increase in length results in a decrease in natural frequencies. Stability analysis is conducted, and flutter and divergence bifurcations are accurately studied. The results show that the shell with clamped boundary conditions is more stable than the supported boundary condition due to its higher bending stiffness. As the distance between the layers of the fluid-conveying nano-spiral shell increases, instability occurs at a lower fluid velocity. Conversely, as the thickness of the shell increases, the nanostructure becomes unstable at higher fluid velocities.

Charge-Density-Wave Control by Adatom Manipulation and Its Effect on Magnetic Nanostructures.

Charge-density waves (CDWs) are correlated states of matter, in which the electronic density is modulated periodically due to electronic and phononic interactions. Often, CDW phases coexist with other correlated states, such as superconductivity, spin-density waves, or Mott insulators. Controlling CDW phases may, therefore, enable the manipulation of the energy landscape of these interacting states. The transition metal dichalcogenide 2H-NbSe2 hosts both CDW order and superconductivity, with the incommensurate CDW phase resulting in different CDW-to-lattice alignments at the atomic scale. Using scanning tunneling microscopy, we position adatoms on the surface to induce reversible CDW domain switching. We show that the domain structure critically affects other local interactions, particularly the hybridization of Yu-Shiba-Rusinov states, which emerge from exchange interactions of magnetic Fe atoms with the superconductor. Our results suggest that CDW manipulation could also be used to introduce domain walls into coupled spin chains on superconductors, potentially impacting topological superconductivity.