Journal of Chemical Theory and Computation最新文献

Mechanisms of anion permeation within ion channels and nanopores remain poorly understood. Recent cryo-electron microscopy structures of the human bestrophin 1 Cl^- channel (hBest1) provide an opportunity to evaluate ion interactions predicted by molecular dynamics (MD) simulations against experimental observations. Here, we implement the fully polarizable force field AMOEBA in MD simulations on different conformations of hBest1. This force field models multipole moments up to the quadrupole. Using this approach, we model key biophysical properties of the channel that can only be simulated when electronic polarization is included in the molecular models and show that Cl^- permeation through the neck of the pore is achieved through hydrophobic solvation concomitant with partial ion dehydration. Furthermore, we demonstrate how such polarizable simulations can help determine the identity of ion-like densities within high-resolution cryo-EM structures and demonstrate that neglecting polarization places Cl^- at positions that do not correspond to their experimentally resolved location. Overall, our results demonstrate the importance of including electronic polarization in realistic and physically accurate models of biological systems, especially channels and pores that selectively permeate anions.

Molecular dynamics is a popular method for evaluating the tensile stress behaviors of many nanomaterials; however, few manuscripts include their thermostat and barostat damping parameters along with their methods. Here, we illustrate the demonstrable effect that barostat integration has on system dynamics during uniaxial testing under a Nosé-Hoover scheme. Three systems are tested: a 2D graphene sheet, a 3D continuous aluminum volume, and a 3D discontinuous polyvinyl alcohol volume. Isobaric-isothermal (NPT) and isochoric-isothermal (NVT) integration methods are intrinsically related, with high NPT barostat damping parameters converging on the NVT system. As a result, slight variations in barostat damping parameters may elicit different stress responses with higher damping parameters directly linked to increased levels of triaxial strain.

Electron density is a fundamental quantity that can in principle determine all ground state electronic properties of a given system. Although machine learning (ML) models for electron density based on either an atom-centered basis or a real-space grid have been proposed, the demand for a number of high-order basis functions or grid points is enormous. In this work, we propose an efficient grid-point sampling strategy that combines targeted sampling favoring a large density and a screening of grid points associated with linearly independent atomic features. This new sampling strategy is integrated with a field-induced recursively embedded atom neural network model to develop a real-space grid-based ML model for the electron density and its response to an electric field. This approach is applied to a QM9 molecular data set, a H₂O/Pt(111) interfacial system, an Au(100) electrode, and an Au nanoparticle under an electric field. The number of training points is found to be much smaller than previous models, while yielding comparably accurate predictions for the electron density of the entire grid. The resultant machine-learned electron density model enables us to properly partition partial charge onto each atom and analyze the charge variation upon proton transfer in the H₂O/Pt(111) system. The machine-learning electronic response model allows us to predict charge transfer and the electrostatic potential change induced by an electric field applied to an Au(100) electrode or an Au nanoparticle.

The charge transfer (CT) reactions in nucleic acids are crucial for genome damage and repair and nanoelectronics using DNA as a molecular conductor. Previous experimental and theoretical works underlined the significance of nucleic acid structural dynamics on CT kinetics, requiring models that incorporate the dynamics of the nucleic acid, solvents, and counterions. Here, we investigated hole transfer kinetics in poly adenine single and double strands at various temperatures and the rate enhancement due to adenine-to-7-deazaadenine mutation by means of a QM/MM approach. We found that the hole transfer rate in poly adenine double strands increases with temperature while the helix conformation is retained, whereas single strands exhibit the opposite thermal response. Additionally, the positive charge migrates more efficiently in poly-7-deazaadenine double strands. Our results, consistent with experimental data, suggest that a thermally induced hopping model can accurately describe CT kinetics in these sequences. The approach is transferable for studying CT reactions in other nucleic acid strands.

The assessment of electronic structure descriptions utilized in the simulation of the ultrafast excited-state dynamics of Fe(II) complexes is presented. Herein, we evaluate the performance of the RPBE, OPBE, BLYP, B3LYP, B3LYP*, PBE0, TPSSh, CAM-B3LYP, and LC-BLYP (time-dependent) density functional theory (DFT/TD-DFT) methods in full-dimensional trajectory surface hopping (TSH) simulations carried out on linear vibronic coupling (LVC) potentials. We exploit the existence of time-resolved X-ray emission spectroscopy (XES) data for the [Fe(bmip)₂]²⁺ and [Fe(terpy)₂]²⁺ prototypes for dynamics between metal-to-ligand charge-transfer (MLCT) and metal-centered (MC) states, which serve as a reference to benchmark the calculations (bmip = 2,6-bis(3-methyl-imidazole-1-ylidine)-pyridine, terpy = 2,2':6',2″-terpyridine). The results show that the simulated ultrafast population dynamics between MLCT and MC states with various spin multiplicities (singlet, triplet, and quintet) highly depend on the utilized DFT/TD-DFT method, with the percentage of exact (Hartree-Fock) exchange being the governing factor. Importantly, B3LYP* and TPSSh are the only DFT/TD-DFT methods with satisfactory performance, best reproducing the experimentally resolved dynamics for both complexes, signaling an optimal balance in the description of MLCT-MC energetics. This work demonstrates the power of combining TSH/LVC dynamics simulations with time-resolved experimental reference data to benchmark full-dimensional potential energy surfaces.

The development of machine-learning (ML) potentials offers significant accuracy improvements compared to molecular mechanics (MM) because of the inclusion of quantum-mechanical effects in molecular interactions. However, ML simulations are several times more computationally demanding than MM simulations, so there is a trade-off between speed and accuracy. One possible compromise are hybrid machine learning/molecular mechanics (ML/MM) approaches with mechanical embedding that treat the intramolecular interactions of the ligand at the ML level and the protein-ligand interactions at the MM level. Recent studies have reported improved protein-ligand binding free energy results based on ML/MM using ANI-2x with mechanical embedding, arguing that intramolecular interactions like torsion potentials of the ligand are often the limiting factor for accuracy. This claim is evaluated based on 108 relative binding free energy calculations for four different benchmark systems. As an alternative strategy, we also tested a tool that fits the MM dihedral potentials to the ML level of theory. Fitting was performed with the ML potentials ANI-2x and AIMNet2, and, for the benchmark system TYK2, also with quantum-mechanical calculations using ωB97M-D3(BJ)/def2-TZVPPD. Overall, the relative binding free energy results from MM with Open Force Field 2.2.0, MM with ML-fitted torsion potentials, and the corresponding ML/MM end-state corrected simulations show no statistically significant differences in the mean absolute errors (between 0.8 and 0.9 kcal mol^-1). This can probably be explained by the usage of the same MM parameters to calculate the protein-ligand interactions. Therefore, a well-parametrized force field is on a par with simple mechanical embedding ML/MM simulations for protein-ligand binding. In terms of computational costs, the reparametrization of poor torsional potentials is preferable over employing computationally intensive ML/MM simulations of protein-ligand complexes with mechanical embedding. Also, the refitting strategy leads to lower variances of the protein-ligand binding free energy results than the ML/MM end-state corrections. For free energy corrections with ML/MM, the results indicate that better convergence and more advanced ML/MM schemes will be required for applications in computer-guided drug discovery.

Machine learning (ML) methods offer a promising route to the construction of universal molecular potentials with high accuracy and low computational cost. It is becoming evident that integrating physical principles into these models, or utilizing them in a Δ-ML scheme, significantly enhances their robustness and transferability. This paper introduces PM6-ML, a Δ-ML method that synergizes the semiempirical quantum-mechanical (SQM) method PM6 with a state-of-the-art ML potential applied as a universal correction. The method demonstrates superior performance over standalone SQM and ML approaches and covers a broader chemical space than its predecessors. It is scalable to systems with thousands of atoms, which makes it applicable to large biomolecular systems. Extensive benchmarking confirms PM6-ML's accuracy and robustness. Its practical application is facilitated by a direct interface to MOPAC. The code and parameters are available at https://github.com/Honza-R/mopac-ml.

The transition from B-DNA to A-DNA occurs in many protein-DNA interactions or in DNA/RNA hybrid duplexes, and thus plays a role in many important biomolecular processes that convey the biological function of DNA. However, the stability of A-DNA is severely underestimated in current AMBER force fields such as OL15, OL21 or bsc1, potentially leading to unstable or deformed protein-DNA complexes. In this study, we refine the deoxyribose dihedral potential to increase the stability of the north (N) puckering present in A-DNA. The new parameters, termed OL24, model A/B equilibrium in B-DNA duplexes in water in good agreement with nuclear magnetic resonance (NMR) experiment. They also improve the description of DNA/RNA hybrids and the transition of the DNA duplex to the A-form in concentrated ethanol solutions. These refinements significantly improve the modeling of protein-DNA complexes, increasing their structural stability and A-form population, while maintaining accurate representation of canonical B-DNA duplexes. Overall, the new parameters should allow more reliable modeling of the thermodynamic equilibrium between A- and B-DNA forms and the interactions of DNA with proteins.

With the advancement of high harmonic generation and X-ray free-electron lasers (XFELs) to the attosecond domain, the studies of the ultrafast electron and spin dynamics became possible. Yet, the methods for efficient control and measurement of the quantum state are to be further developed. In this publication, we propose using magnetic X-ray scattering (MXS) for resolving the molecular spin-state dynamics and establish a complete protocol to simulate MXS diffraction patterns in molecules with ab initio quantum chemistry based on the multiconfigurational method. The performance of the method is demonstrated for the simulation of the spin-flip dynamics in the TiCl₄ molecule, initiated by an ultrashort X-ray pulse. The consistent variation of the electron population and the circular dichroic patterns show the capability of MXS to quantitatively detect the spin-state dynamics in real time quantitatively. We also conclude that the spatial shape and extent of the spin density can also be inferred by analyzing the diffraction patterns for randomly oriented and aligned molecules.