What if you could tell, before someone got a vaccine, whether it would work well for them? Researchers report an AI model that does exactly that — by reading the antibodies a person already carries.
A team at Arizona State University’s Biodesign Institute, led by Joshua LaBaer, trained a deep-learning model on antibody “fingerprints” across 185 antigens (from viruses, bacteria and autoimmune targets) to identify “sentinel antibodies” — markers that reflect a person’s overall antibody-producing capacity rather than fighting the vaccine target directly. The study analyzed 8,687 blood samples from 4,089 people (2,445 healthy and 1,644 immunosuppressed), focused on COVID-19 vaccination. It was published in Cell.
What the AI saw
Higher pre-existing antibodies to certain common microbes — including Staphylococcus aureus, respiratory syncytial virus (RSV) and a human respirovirus — correlated with stronger COVID-19 vaccine responses. In other words, a baseline snapshot of the immune system helped predict who would respond robustly and who might not.
Why it matters
The approach could help identify patients who need extra doses or alternative protection, guide vaccine development, and personalize care for immunocompromised people who often respond poorly. “Certain biomarkers, when analyzed with AI, can predict who is likely to respond well to a vaccine, even before they receive it,” the researchers said. It is a research finding for now, but a practical step toward more personalized immunization.