polyGen: A Learning Framework for Atomic-level Polymer Structure Generation
This work addresses a bottleneck in polymer informatics for researchers and industries in energy, electronics, and medical sectors by enabling on-demand structure generation, though it is a first proof-of-concept and thus incremental in establishing a new capability.
The paper tackles the problem of generating realistic 3D atomic structures for synthetic polymers, which is challenging due to representation and dataset constraints, and introduces polyGen as the first generative model for this task, achieving promising performance in generating diverse conformations from minimal inputs like repeat unit chemistry.
Synthetic polymeric materials underpin fundamental technologies in the energy, electronics, consumer goods, and medical sectors, yet their development still suffers from prolonged design timelines. Although polymer informatics tools have supported speedup, polymer simulation protocols continue to face significant challenges in the on-demand generation of realistic 3D atomic structures that respect conformational diversity. Generative algorithms for 3D structures of inorganic crystals, bio-polymers, and small molecules exist, but have not addressed synthetic polymers because of challenges in representation and dataset constraints. In this work, we introduce polyGen, the first generative model designed specifically for polymer structures from minimal inputs such as the repeat unit chemistry alone. polyGen combines graph-based encodings with a latent diffusion transformer using positional biased attention for realistic conformation generation. Given the limited dataset of 3,855 DFT-optimized polymer structures, we incorporate joint training with small molecule data to enhance generation quality. We also establish structure matching criteria to benchmark our approach on this novel problem. polyGen overcomes the limitations of traditional crystal structure prediction methods for polymers, successfully generating realistic and diverse linear and branched conformations, with promising performance even on challenging large repeat units. As the first atomic-level proof-of-concept capturing intrinsic polymer flexibility, it marks a new capability in material structure generation.