Completely Multipolar Model for Many-Body Water-Ion and Ion-Ion Interactions.

This work constructs an advanced force field, the Completely Multipolar Model (CMM), to quantitatively reproduce each term of an energy decomposition analysis (EDA) for aqueous solvated alkali metal cations and halide anions and their ion pairings. We find that all individual EDA terms remain well-approximated in the CMM for ion-water and ion-ion interactions, except for polarization, which shows errors due to the partial covalency of ion interactions near their equilibrium. We quantify the onset of the dative bonding regime by examining the change in molecular polarizability and Mayer bond indices as a function of distance, showing that partial covalency manifests by breaking the symmetry of atomic polarizabilities while strongly damping them at short-range.

Second-Generation Energy Decomposition Analysis of Intermolecular Interaction Energies from the Second-Order Mo̷ller-Plesset Theory: An Extensible, Orthogonal Formulation with Useful Basis Set Convergence for All Terms.

Energy decomposition analysis (EDA) based on density functional theory (DFT) and self-consistent field (SCF) calculations has become widely used for understanding intermolecular interactions. This work reports a new approach to EDA for post-SCF wave functions based on closed-shell restricted second-order Mo̷ller-Plesset (MP2) together with an efficient implementation that generalizes the successful SCF-level second-generation absolutely localized molecular orbital EDA approach, ALMO-EDA-II, and improves upon MP2 ALMO-EDA-I. The new MP2 ALMO-EDA-II provides distinct energy contributions for a frozen interaction energy containing permanent electrostatics and Pauli repulsions, polarized energy-yielding induced electrostatics, dispersion-corrected energy, and the fully relaxed energy, which describes charge transfer.

UCBShift 2.0: Bridging the Gap from Backbone to Side Chain Protein Chemical Shift Prediction for Protein Structures.

Chemical shifts are a readily obtainable NMR observable that can be measured with high accuracy, and because they are sensitive to conformational averages and the local molecular environment, they yield detailed information about protein structure in solution. To predict chemical shifts of protein structures, we introduced the UCBShift method that uniquely fuses a transfer prediction module, which employs sequence and structure alignments to select reference chemical shifts from an experimental database, with a machine learning model that uses carefully curated and physics-inspired features derived from X-ray crystal structures to predict backbone chemical shifts for proteins. In this work, we extend the UCBShift 1.

Near equivalence of polarizability and bond order flux metrics for describing covalent bond rearrangements.

Identification of the breaking point for the chemical bond is essential for our understanding of chemical reactivity. The current consensus is that a point of maximal electron delocalization along the bonding axis separates the different bonding regimes of reactants and products. This maximum transition point has been investigated previously through the total position spread and the bond-parallel components of the static polarizability tensor for describing covalent bond breaking.

Analytical ab initio hessian from a deep learning potential for transition state optimization.

Identifying transition states-saddle points on the potential energy surface connecting reactant and product minima-is central to predicting kinetic barriers and understanding chemical reaction mechanisms. In this work, we train a fully differentiable equivariant neural network potential, NewtonNet, on thousands of organic reactions and derive the analytical Hessians. By reducing the computational cost by several orders of magnitude relative to the density functional theory (DFT) ab initio source, we can afford to use the learned Hessians at every step for the saddle point optimizations.