AI-Designed Mutation-Resistant Broad Neutralization Antibodies Against Multiple SARS-CoV-2 Strains

Artificial intelligence has the advantage of allowing for the analysis of big and ever-changing data such as the dynamic genome information of a constantly evolving virus. In this study, we have applied AI technologies in the computational design of broad neutralization antibodies against over 1300 different historical SARS-CoV-2 strains with very small computational cost. The AI-designed antibodies were tested in vitro and demonstrated high potency against multiple strains in pseudo-virus and real virus neutralization assays. These AI-designed antibodies also exhibited a very high cross-binding hit rate against different RBD mutants in ELISA assays. The results shed light on future therapeutic designs for pandemic preparedness. The study also indicates that there are hidden patterns in viral evolution and that these patterns can be learned by AI to design antibodies against current and future mutant strains within certain evolutionary windows. 

 Key Findings: 

  • AI can allow for the design antibodies with in vitro efficacy and drastically reduce the time and cost of antibody engineering such as affinity maturation. 
  • AI can allow for the design of cross-binding antibodies against a large number of different antigen population such as viral mutant strains. 
  • AI can reveal patterns of viral evolution processes and allow for the design of antibodies against future viruses beyond current strains.