An End-to-End Deep Generative Pipeline for De Novo Antigen Discovery in Epitope-Based Viral Vaccines

Overview

Rapid viral evolution continues to outpace conventional vaccine development, particularly for highly variable pathogens such as Influenza A, Norovirus, and HIV. Although epitope-based vaccines help focus immune responses towards target regions on viral proteins, their design remains heuristic and difficult to scale. No unified system currently integrates epitope optimization, global HLA population coverage analysis, rational antigen architecture engineering, and immune validation within a single automated framework, hindering efficient development of universal vaccines in silico. This project pioneers a fully automated, end-to-end pipeline for de novo epitope-based antigen discovery to enable universal-targeting antigen construction. Using a novel multi-objective optimization framework, the platform identifies highly conserved, high-immunogenicity epitopes with strong predicted MHC-I binding affinity and low sequence variability quantified by Shannon entropy. Optimized epitope panels are assembled into structured antigen constructs, refined via CMA-ES optimization, and evaluated using computational immune simulations to quantify antibody kinetics, cellular activation, and memory formation. Applied to Influenza A, Norovirus, and HIV, the pipeline generated novel multi-epitope antigens optimized for high conservation (mean Shannon entropy < 0.30), strong MHC-I binding (mean predicted affinity <100 nM), and broad global population coverage (92%). Generated antigens also demonstrated favorable structural stability and low toxicity. Simulated immune responses using C-ImmSim predicted rapid antigen clearance, robust IgG class switching, and sustained memory B- and CD4+ T-cell populations, supporting the strain-transcending immunogenic potential of the generated antigens. This work therefore establishes a novel, fully automated framework for de novo epitope-based antigen engineering, enabling rapid and scalable development of universal vaccine candidates.

Competition history

• CSEF 2026 Biochemistry/ Molecular Biology (Senior Division) · Entry S-04-23

Related projects

Closest projects by meaning, across every fair and year in the corpus.

Browse more like this

• All Cellular and Molecular Biology projects
• CSEF Cellular and Molecular Biology projects
• Every CSEF 2026 project