Constrained Submodular Optimization for Vaccine Design

Advances in machine learning have enabled the prediction of immune system responses to prophylactic and therapeutic vaccines. However, engineering task designing vaccines remains a challenge. In particular, genetic variability human makes it difficult design peptide that provide widespread immunity vaccinated populations. We introduce framework for evaluating uses probabilistic models, demonstrate its ability produce designs SARS-CoV-2 vaccine outperform previous designs. theoretical analysis approximability, scalability, complexity our framework.