Leveraging symmetry to predict self-assembly of multiple polymers

Protein self-assembly is fundamental to biological function and disease. Experimentally, the atomic-level structure is difficult to obtain and the assembly mechanism is poorly understood. The large number of possible states accessible to such systems limits computational prediction. Here, I introduce a new computational approach that enforces conformational symmetry, whereby all chains in the system adopt the same conformation. Using this approach on a 2D lattice, a designed multi-chain conformation is found more than four orders of magnitude faster than existing approaches. Furthermore, the free energy landscape can be efficiently computed, showing potential for enabling atomistic prediction of protein self-assembly.