Exploring the frontiers of condensed-phase chemistry with a general reactive machine learning potential.

Atomistic simulation has a broad range of applications from drug design to materials discovery. Machine learning interatomic potentials (MLIPs) have become an efficient alternative to computationally expensive ab initio simulations. For this reason, chemistry and materials science would greatly benefit from a general reactive MLIP, that is, an MLIP that is applicable to a broad range of reactive chemistry without the need for refitting. Here we develop a general reactive MLIP (ANI-1xnr) through automated sampling of condensed-phase reactions. ANI-1xnr is then applied to study five distinct systems: carbon solid-phase nucleation, graphene ring formation from acetylene, biofuel additives, combustion of methane and the spontaneous formation of glycine from early earth small molecules. In all studies, ANI-1xnr closely matches experiment (when available) and/or previous studies using traditional model chemistry methods. As such, ANI-1xnr proves to be a highly general reactive MLIP for C, H, N and O elements in the condensed phase, enabling high-throughput in silico reactive chemistry experimentation.

Teaching a neural network to attach and detach electrons from molecules.

Interatomic potentials derived with Machine Learning algorithms such as Deep-Neural Networks (DNNs), achieve the accuracy of high-fidelity quantum mechanical (QM) methods in areas traditionally dominated by empirical force fields and allow performing massive simulations. Most DNN potentials were parametrized for neutral molecules or closed-shell ions due to architectural limitations. In this work, we propose an improved machine learning framework for simulating open-shell anions and cations. We introduce the AIMNet-NSE (Neural Spin Equilibration) architecture, which can predict molecular energies for an arbitrary combination of molecular charge and spin multiplicity with errors of about 2-3 kcal/mol and spin-charges with error errors ~0.01e for small and medium-sized organic molecules, compared to the reference QM simulations. The AIMNet-NSE model allows to fully bypass QM calculations and derive the ionization potential, electron affinity, and conceptual Density Functional Theory quantities like electronegativity, hardness, and condensed Fukui functions. We show that these descriptors, along with learned atomic representations, could be used to model chemical reactivity through an example of regioselectivity in electrophilic aromatic substitution reactions.

Approaching coupled cluster accuracy with a general-purpose neural network potential through transfer learning.

Computational modeling of chemical and biological systems at atomic resolution is a crucial tool in the chemist's toolset. The use of computer simulations requires a balance between cost and accuracy: quantum-mechanical methods provide high accuracy but are computationally expensive and scale poorly to large systems, while classical force fields are cheap and scalable, but lack transferability to new systems. Machine learning can be used to achieve the best of both approaches. Here we train a general-purpose neural network potential (ANI-1ccx) that approaches CCSD(T)/CBS accuracy on benchmarks for reaction thermochemistry, isomerization, and drug-like molecular torsions. This is achieved by training a network to DFT data then using transfer learning techniques to retrain on a dataset of gold standard QM calculations (CCSD(T)/CBS) that optimally spans chemical space. The resulting potential is broadly applicable to materials science, biology, and chemistry, and billions of times faster than CCSD(T)/CBS calculations.

Discovering a Transferable Charge Assignment Model Using Machine Learning.

Partial atomic charge assignment is of immense practical value to force field parametrization, molecular docking, and cheminformatics. Machine learning has emerged as a powerful tool for modeling chemistry at unprecedented computational speeds given accurate reference data. However, certain tasks, such as charge assignment, do not have a unique solution. Herein, we use a machine learning algorithm to discover a new charge assignment model by learning to replicate molecular dipole moments across a large, diverse set of nonequilibrium conformations of molecules containing C, H, N, and O atoms. The new model, called Affordable Charge Assignment (ACA), is computationally inexpensive and predicts dipoles of out-of-sample molecules accurately. Furthermore, dipole-inferred ACA charges are transferable to dipole and even quadrupole moments of much larger molecules than those used for training. We apply ACA to dynamical trajectories of biomolecules and produce their infrared spectra. Additionally, we find that ACA assigns similar charges to Charge Model 5 but with greatly reduced computational cost.

Photoexcited Nonadiabatic Dynamics of Solvated Push-Pull π-Conjugated Oligomers with the NEXMD Software.

Solvation can be modeled implicitly by embedding the solute in a dielectric cavity. This approach models the induced surface charge density at the solute-solvent boundary, giving rise to extra Coulombic interactions. Herein, the Nonadiabatic EXcited-state Molecular Dynamics (NEXMD) software was used to model the photoexcited nonradiative relaxation dynamics in a set of substituted donor-acceptor oligo( p-phenylenevinylene) (PPVO) derivatives in the presence of implicit solvent. Several properties of interest including optical spectra, excited state lifetimes, exciton localization, excited state dipole moments, and structural relaxation are calculated to elucidate dependence of functionalization and solvent polarity on photoinduced nonadiabatic dynamics. Results show that solvation generally affects all these properties, where the magnitude of these effects vary from one system to another depending on donor-acceptor substituents and molecular polarizability. We conclude that implicit solvation can be directly incorporated into nonadiabatic simulations within the NEXMD framework with little computational overhead and that it qualitatively reproduces solvent-dependent effects observed in solution-based spectroscopic experiments.

Design principles from multiscale simulations to predict nanostructure in self-assembling ionic liquids.

Molecular dynamics simulations (up to the nanoscale) were performed on the 3-methyl-1-pentylimidazolium ionic liquid cation paired with three anions; chloride, nitrate, and thiocyanate as aqueous mixtures, using the effective fragment potential (EFP) method, a computationally inexpensive way of modeling intermolecular interactions. The simulations provided insight (preferred geometries, radial distribution functions and theoretical proton NMR resonances) into the interactions within the ionic domain and are validated against 1H NMR spectroscopy and small- and wide-angle X-ray scattering experiments on 1-decyl-3-methylimidazolium. Ionic liquids containing thiocyanate typically resist gelation and form poorly ordered lamellar structures upon mixing with water. Conversely, chloride, a strongly coordinating anion, normally forms strong physical gels and produces well-ordered nanostructures adopting a variety of structural motifs over a very wide range of water compositions. Nitrate is intermediate in character, whereby upon dispersal in water it displays a range of viscosities and self-assembles into nanostructures with considerable variability in the fidelity of ordering and symmetry, as a function of water content in the binary mixtures. The observed changes in the macro and nanoscale characteristics were directly correlated to ionic domain structures and intermolecular interactions as theoretically predicted by the analysis of MD trajectories and calculated RDFs. Specifically, both chloride and nitrate are positioned in the plane of the cation. Anion to cation proximity is dependent on water content. Thiocyanate is more susceptible to water insertion into the second solvent shell. Experimental 1H NMR chemical shifts monitor the site-specific competition dependence with water content in the binary mixtures. Thiocyanate preferentially sits above and below the aromatic ring plane, a state disallowing interaction with the protons on the imidazolium ring.

Dynamics of a myoglobin mutant enzyme: 2D IR vibrational echo experiments and simulations.

Myoglobin (Mb) double mutant T67R/S92D displays peroxidase enzymatic activity in contrast to the wild type protein. The CO adduct of T67R/S92D shows two CO absorption bands corresponding to the A(1) and A(3) substates. The equilibrium protein dynamics for the two distinct substates of the Mb double mutant are investigated by using two-dimensional infrared (2D IR) vibrational echo spectroscopy and molecular dynamics (MD) simulations. The time-dependent changes in the 2D IR vibrational echo line shapes for both of the substates are analyzed using the center line slope (CLS) method to obtain the frequency-frequency correlation function (FFCF). The results for the double mutant are compared to those from the wild type Mb. The experimentally determined FFCF is compared to the FFCF obtained from molecular dynamics simulations, thereby testing the capacity of a force field to determine the amplitudes and time scales of protein structural fluctuations on fast time scales. The results provide insights into the nature of the energy landscape around the free energy minimum of the folded protein structure.