Scientists have used artificial intelligence to understand why the body’s response to vaccines can vary significantly between countries and among different groups of people. A new study, conducted by an international team led by researchers from the University of York, showed that differences in immune response are linked to a complex combination of biological and social factors.

The results were published in the journal Cell.

Researchers applied machine learning methods to analyze large-scale immunological data. The algorithms examined immune system indicators in various individuals, including both healthy participants and people living with HIV.

The models were able to identify subtle differences in how the body responds to vaccines, as well as detect so-called “outliers”—cases in which the immune response deviated significantly from expectations.

The analysis revealed that vaccine effectiveness is influenced by multiple factors, including age, genetic characteristics, the presence of chronic diseases, and overall immune system health. Even when using the same vaccine, people may produce different levels of antibodies and develop varying degrees of immune protection.

According to the researchers, these differences help explain why vaccination can show varying effectiveness in different regions of the world. In addition to the biological characteristics of populations, environmental conditions, prior infections, nutrition, and access to healthcare may also play a role.

Artificial intelligence allows researchers to account for many such variables simultaneously and to identify hidden patterns that are difficult to detect using traditional analytical methods.

The scientists believe that these findings could be an important step toward the development of personalized vaccination. In the future, such models could help predict how an individual or an entire population will respond to a specific vaccine. This would enable more precise immunization strategies, improve vaccine design, and enhance effectiveness across countries and diverse population groups.