Automatic chemical design using a data-driven continuous representation of molecules
This addresses the challenge of efficient chemical design for researchers in drug discovery and materials science, representing a novel method for a known bottleneck.
The authors tackled the problem of generating and optimizing novel chemical compounds by developing a method to convert discrete molecular representations into a continuous latent space, enabling efficient exploration and optimization through operations like random decoding and gradient-based search. They demonstrated this approach on drug-like molecules and small molecules, showing it can automatically generate new structures.
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 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 the set of molecules with fewer that nine heavy atoms.