3.23.21-Health-David-Gifford

• Video details

  We present combinatorial machine learning methods to evaluate and optimize peptide vaccine formulations using a solution to the maximum n-times coverage problem.   We apply these new methods to design a peptide vaccine to induce cellular immunity to SARS-CoV-2, and show that our solution is superior to 29 other published COVID-19 peptide vaccine designs in predicted population coverage and the expected number of peptides displayed by each individual's HLA molecules.   Our proposed SARS-CoV-2 MHC class I vaccine formulations provide 99.99% predicted population coverage with at least one vaccine peptide-HLA average hit per person with all vaccine peptides perfectly conserved across 4,690 geographically sampled SARS-CoV-2 genomes.  Our proposed MHC class II vaccine formulations provide 95.62% predicted coverage with at least one vaccine peptide-HLA average hits per person with all peptides having an observed mutation probability of ≤ 0.001.  Our population coverage estimates integrate clinical data on peptide immunogenicity in convalescent COVID-19 patients and machine learning predictions.   Our use of conserved viral sequences in vaccine designs is intended to make our vaccines effective against new viral strains.   We are presently testing our vaccine designs in animal models.

Locked Interactive transcript