Antibody design across observed and forecasted antigenic evolution

Germinal Multi designs and rescues antibodies that jointly optimize one paratope for multiple antigenic variants, yielding breadth against resistance-associated EGFR variants and distinct historical SARS-CoV-2 variants.

• Why it matters: Antibody therapies can lose efficacy as targets evolve, and designing binding interfaces that tolerate antigenic change—especially future escape—is difficult.
• What they did: The framework performs de novo antibody design and rescue by multi-target optimization across observed antigenic variation and prospective antibody escape, then screens tens of designs per campaign.
• The result: It produced antibodies that recognized variants withheld from design and, for BD55-1205, restored neutralization of a prospective escape variant engineered using evolutionary forecasting while preserving activity against the parental variant.

AI summary of the abstract · may contain errors · not medical advice · read the paper.