The development of efficient models for predicting specific properties through machine learning is of great importance for the innovation of chemistry and material science. However, predicting global electronic structure properties like Frontier molecular orbital highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO) energy levels and their HOMO–LUMO gaps from the small-sized molecule data to larger molecules remains a challenge. Here, we develop a multilevel attention neural network, named DeepMoleNet, to enable chemical interpretable insights being fused into multitask learning through (1) weighting contributions from various atoms and (2) taking the atom-centered symmetry functions (ACSFs) as the teacher descriptor. The efficient prediction of 12 properties including dipole moment, HOMO, and Gibbs free energy within chemical accuracy is achieved by using multiple benchmarks, both at the equilibrium and nonequilibrium geometries, including up to 110,000 records of data in QM9, 400,000 records in MD17, and 280,000 records in ANI-1ccx for random split evaluation. The good transferability for predicting larger molecules outside the training set is demonstrated in both equilibrium QM9 and Alchemy data sets at the density functional theory (DFT) level. Additional tests on nonequilibrium molecular conformations from DFT-based MD17 data set and ANI-1ccx data set with coupled cluster accuracy as well as the public test sets of singlet fission molecules, biomolecules, long oligomers, and protein with up to 140 atoms show reasonable predictions for thermodynamics and electronic structure properties. The proposed multilevel attention neural network is applicable to high-throughput screening of numerous chemical species in both equilibrium and nonequilibrium molecular spaces to accelerate rational designs of drug-like molecules, material candidates, and chemical reactions.

We have implemented the calculations of NMR parameters within the generalized energy-based fragmentation (GEBF) method for condensed-phase systems with periodic boundary conditions (PBC). In this PBC-GEBF approach, NMR parameters of molecules in a unit cell are assembled as a linear combination of the corresponding quantities from a series of small embedded subsystems. To treat condensed-phase systems containing large molecules, we propose a novel “fragment-based” strategy for building subsystems, while our previously reported “molecule-based” strategy for construction of subsystems is appropriate for periodic systems with small molecules. The “fragment-based” strategy in PBC-GEBF is demonstrated to be much more efficient than its “molecule-based” counterpart to treat crystals of large molecules. With the “molecule-based” PBC-GEBF method, we obtained consistently good NMR parameters of liquid water with B3LYP on top of neural-network-potential-based ab initio molecular dynamics (AIMD) snapshots. With the “fragment-based” PBC-GEBF approach, we predicted the 1H chemical shifts of a large macrocycle in solution based on a series of classical MD snapshots. The calculated results are in good accord with the experimental chemical shifts. Therefore, the PBC-GEBF method is expected to be a reliable and efficient tool for predicting NMR parameters of large complex systems in solutions.

An on-the-fly fragment-based machine learning (ML) approach was developed to construct the machine learning force field for large complex systems. In this approach, the energy, forces, and molecular properties of the target system are obtained by combining machine learning force fields of various subsystems with the generalized energy-based fragmentation (GEBF) approach. Using nonparametric Gaussian process (GP) model, all the force fields of subsystems are automatically generated online without data selection and parameter optimization. With the GEBF-ML force field constructed for a normal alkane, C60H122, long-time molecular dynamics (MD) simulations are performed on different sizes of alkanes, and the predicted energy, forces, and molecular properties (dipole moment) are favorably comparable with full quantum mechanics (QM) calculations. The predicted IR spectra also show excellent agreement with the direct ab initio MD results. Our results demonstrate that the GEBF-ML method provides an automatic and efficient way to build force fields for a broad range of complex systems such as biomolecules and supramolecular systems.

Electrochemical organic synthesis has attracted increasing attentions as a sustainable and versatile synthetic platform. Quantitative assessment of the electro‐organic reactions, including reaction thermodynamics, electro‐kinetics, and coupled chemical processes, can lead to effective analytical tool to guide their future design. Herein, we demonstrate that electrochemical parameters such as onset potential, Tafel slope, and effective voltage can be utilized as electro‐descriptors for the evaluation of reaction conditions and prediction of reactivities (yields). An “electro‐descriptor‐diagram” is generated, where reactive and non‐reactive conditions/substances show distinct boundary. Successful predictions of reaction outcomes have been demonstrated using electro‐descriptor diagram, or from machine learning algorithms with experimentally‐derived electro‐descriptors. This method represents a promising tool for data‐acquisition, reaction prediction, mechanistic investigation, and high‐throughput screening for general organic electro‐synthesis.

Intermolecular interactions in terms of molecular packing are crucial for the investigation of the absorption spectra of uracil in different environments.