We report a method to convert discrete representations of molecules to and from a multidimensional continuous representation. This model allows us to generate new molecules for efficient exploration and optimization through open-ended spaces of chemical compounds. A deep neural network was trained on hundreds of thousands of existing chemical structures to construct three coupled functions: an encoder, a decoder, and a predictor. The encoder converts the discrete representation of a molecule into a real-valued continuous vector, and the decoder converts these continuous vectors back to discrete molecular representations. The predictor estimates chemical properties from the latent continuous vector representation of the molecule. Continuous representations of molecules allow us to automatically generate novel chemical structures by performing simple operations in the latent space, such as decoding random vectors, perturbing known chemical structures, or interpolating between molecules. Continuous representations also allow the use of powerful gradient-based optimization to efficiently guide the search for optimized functional compounds. We demonstrate our method in the domain of drug-like molecules and also in a set of molecules with fewer that nine heavy atoms.

The discovery of new materials can bring enormous societal and technological progress. In this context, exploring completely the large space of potential materials is computationally intractable. Here, we review methods for achieving inverse design, which aims to discover tailored materials from the starting point of a particular desired functionality. Recent advances from the rapidly growing field of artificial intelligence, mostly from the subfield of machine learning, have resulted in a fertile exchange of ideas, where approaches to inverse molecular design are being proposed and employed at a rapid pace. Among these, deep generative models have been applied to numerous classes of materials: rational design of prospective drugs, synthetic routes to organic compounds, and optimization of photovoltaics and redox flow batteries, as well as a variety of other solid-state materials.

Generative models are becoming a tool of choice for exploring the molecular space. These models learn on a large training dataset and produce novel molecular structures with similar properties. Generated structures can be utilized for virtual screening or training semi-supervized predictive models in the downstream tasks. While there are plenty of generative models, it is unclear how to compare and rank them. In this work, we introduce a benchmarking platform called Molecular Sets (MOSES) to standardize training and comparison of molecular generative models. MOSES provides training and testing datasets, and a set of metrics to evaluate the quality and diversity of generated structures. We have implemented and compared several molecular generation models and suggest to use our results as reference points for further advancements in generative chemistry research. The platform and source code are available at https://github.com/molecularsets/moses.

Sodium-ion batteries have captured widespread attention for grid-scale energy storage owing to the natural abundance of sodium. The performance of such batteries is limited by available electrode materials, especially for sodium-ion layered oxides, motivating the exploration of high compositional diversity. How the composition determines the structural chemistry is decisive for the electrochemical performance but very challenging to predict, especially for complex compositions. We introduce the “cationic potential” that captures the key interactions of layered materials and makes it possible to predict the stacking structures. This is demonstrated through the rational design and preparation of layered electrode materials with improved performance. As the stacking structure determines the functional properties, this methodology offers a solution toward the design of alkali metal layered oxides.

The inverse design of new materials with desired properties is the ultimate goal of materials research, but demonstrating such a possibility for inorganic solid-state materials has been challenging, due partly to the invertibility of representation.Here, we demonstrate that the generative model using invertible image-based representation yields accurate reconstruction performance and can successfully rediscover experimentally known vanadium oxides. The model predicts several completely new compositions and polymorphs of vanadium oxides that are metastable and may be synthesizable.