27 June 2023 Tue 4 pm

In living cells, proteins self-assemble into large-scale functional structures based both on the specific interactions between different spots on their surfaces. The general principles that allow complex interactions to control self-assembly are largely unknown. Here, we introduce model lattice particles with trivial geometries but highly complex interactions and study their self-assembly using numerical simulations. We find that strongly anisotropic particles produce non-trivial aggregates that fall into a few stereotypical categories. Using machine learning, we accurately predict the outcome of this aggregation based on the presence of pair interactions that locally favor periodic un-frustrated arrangements. We also show that the stereotypical aggregates are fixed-points of a numerical real-space renormalization transformation. These results offer a first overview of the rich design space associated with identical particles with complex interactions, and could inspire engineered self-assembling nano-objects as well as help understand the emergence of robust functional protein structures. In particular, we show that local interaction can robustly control the size of spherical or fibrous self-assemblies of identical particle. Finally, we suggest that it is possible to systematically identify the conditions in which proteins would self-assemble into fibrillar aggregate in X-ray crystallographic scattering measurements.

< Previous

Next >