Computational Design of Peptides to Block Binding of the SARS-CoV-2 Spike Protein to Human ACE2

Xiaoqiang Huang; Robin Pearce; Yang Zhang

Based on the constructed protein-peptide complex structure (SARS-CoV-2 RBD/hACE2-22-44G351-357), we performed 1000 independent design trajectories individually, using (1) EvoEF2 18 , a physics-and knowledge-based energy function specifically designed for protein The copyright holder for this preprint (which was not peer-reviewed) is the . https://doi.org/10.1101/2020.03.28.013607 doi: bioRxiv preprint design and (2) a new version of EvoDesign 19 , which combines EvoEF2 and evolutionary profiles for design scoring. A simulated annealing Monte Carlo (SAMC) 27 protocol was used to search for low total energy sequences as previously described 18 . For each trajectory, only the single lowest energy in that design simulation was selected, and therefore 1000 sequences each were collected from the EvoEF2 and EvoDesign designs. The EvoEF2 and EvoDesign designs were separately analyzed to determine the impact of the physics-and profile-based scores. Since SAMC is a stochastic searching method, some of the 1000 sequences were duplicates and thus excluded from analysis. The backbone conformations of the hACE2 peptide and SARS-CoV-2 RBD were held constant during the protein design simulations, all the residues on the peptide were redesigned, and the side-chains of the interface residues on the virus RBD were repacked without design. The non-redundant designed peptides are listed in Supplementary Tables S1-S5, and the raw data are freely available at https://zhanglab.ccmb.med.umich.edu/EvoEF/COVID-19/.