Publications

In drug discovery, molecular dynamics (MD) simulation for protein-ligand binding provides a powerful tool for predicting binding affinities, estimating transport properties, and exploring pocket sites. There has been a long history of improving the efficiency of MD simulations through better numerical methods and, more recently, by utilizing machine learning (ML) methods. Yet, challenges remain, such as accurate modeling of extended-timescale simulations. To address this issue, we propose NeuralMD, the first ML surrogate that can facilitate numerical MD and provide accurate simulations in protein-ligand binding dynamics. We propose a principled approach that incorporates a novel physics-informed multi-grained group symmetric framework. Specifically, we propose (1) the BindingNet model that satisfies group symmetry using vector frames and captures the multi-level protein-ligand interactions, and (2) an augmented neural differential equation solver that learns the trajectory under Newtonian mechanics. For the experiment, we design ten single-trajectory and three multi-trajectory binding simulation tasks. We demonstrate the efficiency and effectiveness of NeuralMD, achieving over 1K$\times$ speedup compared to standard numerical MD simulations. NeuralMD also outperforms all other ML approaches, achieving up to 15$\times$ reduction in reconstruction error and 70% increase in validity. Additionally, we qualitatively illustrate that the oscillations in the predicted trajectories align more closely with ground-truth dynamics than those of other machine-learning methods. We believe NeuralMD paves the foundation for a new research paradigm in simulating protein-ligand dynamics.

The present invention provides for an organic compound comprising electron deficient unit covalently linked to two or more electron rich units. The present invention also provides for a device comprising the organic compound, such as a light-emitting diode, thin-film transistor, chemical biosensor, non-emissive electrochromic, memory device, photovoltaic cells, or the like.

A mild palladium-catalyzed ligand-controlled regioselective 1,3-arylfluorination of 2[<italic>H</italic>]-chromenes has been developed.

The layered P2-K4Co7O14 oxide has been prepared and characterized by means of X-ray diffraction, electrical conductivity, thermopower, and magnetic measurements. The crystal structure of K4Co7O14 (P6(3)/m space group, Z=2, a=7.5171(1) A, and c=12.371(1) A) consists of a stacking of slabs of edge-shared CoO6 octahedra with K+ ions occupying ordered positions in the interslab space, leading to a a0 radical7xa0 radical7 supercell. Potential energy calculations at 0 K are in good agreement with the ordered distribution of potassium ions in the (ab) plane. This oxide is metallic, and the magnetic susceptibility is of Pauli-type, which contrasts with the Curie-Weiss behavior of the homologous NaxCoO2 (x approximately 0.6) oxide with close alkali content. The thermopower at room temperature is about one-third that of polycrystalline Na0.6CoO2.

The development of deterministic models for predicting the accumulation of corrosion damage to zirconium and Zircaloys in pressurized water reactor (PWR) coolant environments requires the acquisition of values for various model parameters. In the present work, the point defect model (PDM) was further developed to account for the properties of passive films comprising hydride barrier layers and porous oxide outer layers that form on zirconium and Zircaloys in high-temperature, hydrogenated aqueous solutions. The model parameter values were extracted from electrochemical impedance spectroscopic data for zirconium in hydrogenated, borate buffer solution at by optimization. The results indicate that the corrosion resistance of zirconium in high-temperature, hydrogenated aqueous solutions is dominated by the outer layer, as was found to be the case for nonhydrogenated solutions where a defective oxide barrier layer forms. The impedance model based on the PDM provides a good account of the growth of the bi-layer passive films described above, and the extracted model parameter values might be used, for example, for predicting the accumulation of general corrosion damage to Zircaloy fuel cladding in PWR operating environments.

Inspired by recent theoretical predictions on band convergence in the tetragonal chalcopyrite compounds, we have explored the influence of the crystal structure on the transport and bandstructure of different quaternary chalcopyrites. In theory, a changing lattice parameter ratio of c/2a towards unity should lead to band convergence due to a more cubic and higher symmetry structure. In accordance with this prediction, the different solid solutions explored in this manuscript show a significant impact on the electronic transport depending on the ratio of the lattice parameters. An increasing lattice parameter ratio results in an increase of the carrier effective masses which can be explained by converging bands, ultimately leading to an increase of the power factor and thermoelectric figure of merit in the class of non-cubic chalcopyrite compounds Cu_2MGeSe_4. However, the calculations via density functional theory show that the critical value of c/2a, where band convergence occurs, will be different from unity due to symmetry and chemical influences on the band structure.

The behavior and role of hydrogen is investigated by using Pt–Ga nano-alloy formation as a probe reaction.