8F53

Top-down design of protein architectures with reinforcement learning

This is a non-PDB format compatible entry.

Summary for 8F53

• Entry DOI: 10.2210/pdb8f53/pdb
• Related: 8F4X
• EMDB information: 28859
• Descriptor: RC_I_2 (1 entity in total)
• Functional Keywords: nanoparticle, capsid, oligomer, de novo design, rosetta, de novo protein, reinforcement learning
• Biological source: synthetic construct
• Total number of polymer chains: 60
• Total formula weight: 353158.14

Authors

Borst, A.J.,Baker, D. (deposition date: 2022-11-11, release date: 2023-05-10, Last modification date: 2024-06-19)

Primary citation

Lutz, I.D.,Wang, S.,Norn, C.,Courbet, A.,Borst, A.J.,Zhao, Y.T.,Dosey, A.,Cao, L.,Xu, J.,Leaf, E.M.,Treichel, C.,Litvicov, P.,Li, Z.,Goodson, A.D.,Rivera-Sanchez, P.,Bratovianu, A.M.,Baek, M.,King, N.P.,Ruohola-Baker, H.,Baker, D.
Top-down design of protein architectures with reinforcement learning.
Science, 380:266-273, 2023

PubMed Abstract: As a result of evolutionary selection, the subunits of naturally occurring protein assemblies often fit together with substantial shape complementarity to generate architectures optimal for function in a manner not achievable by current design approaches. We describe a "top-down" reinforcement learning-based design approach that solves this problem using Monte Carlo tree search to sample protein conformers in the context of an overall architecture and specified functional constraints. Cryo-electron microscopy structures of the designed disk-shaped nanopores and ultracompact icosahedra are very close to the computational models. The icosohedra enable very-high-density display of immunogens and signaling molecules, which potentiates vaccine response and angiogenesis induction. Our approach enables the top-down design of complex protein nanomaterials with desired system properties and demonstrates the power of reinforcement learning in protein design.

PubMed: 37079676
DOI: 10.1126/science.adf6591

Experimental method

ELECTRON MICROSCOPY (2.93 Å)