An educational activity tailored for undergraduate students on AI-driven nanobody design for targeted antiviral therapeutics

Protein-based biologics are converging at an unprecedented pace as generative artificial intelligence (AI) reshapes early-stage drug discovery. Nanobodies, single-domain antibody fragments derived from camelid heavy chain-only immunoglobulins, are an emerging class of antiviral biologics prized for their small size, stability, and access to cryptic viral epitopes. Traditional nanobody discovery relies on animal immunization and phage display screening, a process that can take 6 to 12 months and is entirely inaccessible within a standard undergraduate laboratory session. Here, we describe a succinct, in silico -based laboratory activity that uses a generative AI protein design server, GeoFlow-V3 ( https://prot.design/ ), to allow undergraduate students to design a novel antiviral nanobody against the influenza A virus (IAV) hemagglutinin (HA) protein. This activity builds on our group’s prior use of generative AI for undergraduate siRNA design teaching and extends the same pedagogical model to structural protein engineering. To our knowledge, this is the first use of a generative AI protein design platform for teaching nanobody design at the undergraduate level.