Generative Models for Synthetically Accessible Polymers

N.E. Jackson
University of Illinois Urbana-Champaign,
United States

Keywords: AI, polymer data science


The de novo design of functional polymers is complicated by the vast chemical space and an incomplete understanding of structure-property relations. Recent advances in deep learning have facilitated the efficient exploration of molecular design space, but data sparsity is a major obstacle hindering progress. Here, we introduce the Open Macromolecular Genome (OMG), which contains synthesizable polymer chemistries compatible with known polymerization reactions and commercially available reactants selected for synthetic feasibility. The OMG is used in concert with a synthetically aware generative model to identify property-optimized constitutional repeating units, constituent reactants, and reaction pathways of polymers, thereby advancing polymer design into the realm of synthetic relevance. As a proof-of-principle demonstration, we show that polymers with targeted octanol-water solubilities are readily generated together with monomer reactant building blocks and associated polymerization reactions. Broadly, the OMG is a polymer design approach capable of enabling data intensive generative models for synthetic polymer design.