Machine-Learned Coarse-Grained Models

Optimizing force-field (FF) parameters to perform molecular dynamics (MD) simulations is a challenging and time-consuming process. We present a novel FF optimization framework that integrates MD simulations with particle swarm optimization (PSO) algorithm and artificial neural network (ANN). This new ANN-assisted PSO framework was used to develop transferable coarse-grained (CG) models for D₂O and DMF as a proof of concept. The PSO algorithm was used to generate the set of input FF parameters for the MD simulations of the CG models of these solvents, which were optimized to reproduce their experimental properties. Herein, for the first time, a reverse approach was employed for on-the-fly training of the ANN model, where results (solvent properties) obtained from the MD simulations and their corresponding FF parameters were used as inputs and outputs, respectively. The ANN model was then required to predict a set of new FF parameters, which were tested for their ability to predict the desired experimental properties. This new framework can be extended to integrate any optimization algorithm with ANN and MD simulations to accelerate the FF development.

Neural-Network-Biased Genetic Algorithms for Materials Design: Evolutionary Algorithms That Learn

Machine learning has the potential to dramatically accelerate high-throughput approaches to materials design, as demonstrated by successes in biomolecular design and hard materials design. However, in the search for new soft materials exhibiting properties and performance beyond those previously achieved, machine learning approaches are frequently limited by two shortcomings. First, because they are intrinsically interpolative, they are better suited to the optimization of properties within the known range of accessible behavior than to the discovery of new materials with extremal behavior. Second, they require large pre-existing data sets, which are frequently unavailable and prohibitively expensive to produce. Here we describe a new strategy, the neural-network-biased genetic algorithm (NBGA), for combining genetic algorithms, machine learning, and high-throughput computation or experiment to discover materials with extremal properties in the absence of pre-existing data. Within this strategy, predictions from a progressively constructed artificial neural network are employed to bias the evolution of a genetic algorithm, with fitness evaluations performed via direct simulation or experiment. In effect, this strategy gives the evolutionary algorithm the ability to "learn" and draw inferences from its experience to accelerate the evolutionary process. We test this algorithm against several standard optimization problems and polymer design problems and demonstrate that it matches and typically exceeds the efficiency and reproducibility of standard approaches including a direct-evaluation genetic algorithm and a neural-network-evaluated genetic algorithm. The success of this algorithm in a range of test problems indicates that the NBGA provides a robust strategy for employing informatics-accelerated high-throughput methods to accelerate materials design in the absence of pre-existing data.

A comprehensive QSPR model for dielectric constants of binary solvent mixtures

The dielectric constant is a key physicochemical property in solubility, chemical equilibrium and the synthesis of compounds in pharmaceutical/chemical sciences. In this context, a quantitative structure-property relationship (QSPR) model was designed from 3207 binary solvent mixtures by using 23 calculated experimental-theoretical descriptors including solvent fractions (f₁ and f₂), individual dielectric constants of solvents (dc₁ and dc₂), temperature, and Abraham and Hansen solvation parameters. The QSPR model was developed using a genetic algorithm based multiple linear regression (GA-MLR) and robust regression. Jackknifing was implemented for internal-external validation of the selected descriptors by GA containing f₁, f₂, dc₁ and dc₂. Implementation of jackknifing on the selected descriptors revealed that p values were close to zero. Consequently, the significance of selected descriptors was confirmed through the sign change point of view and their validity was verified. The model was evaluated using the r² and Q²_(F3) parameters as criteria of model prediction ability. The r² values were equal to 0.925 and 0.922, and Q²_(F3) were reported as 0.873 and 0.862 for the cross-validation and prediction steps, respectively. Finally, model performance was clearly acceptable to anticipate the modelling of dielectric constants for a wide range of binary solvent mixtures.