The ability to predict how different individuals will respond to vaccination and to understand what best protects individuals from infection greatly facilitates developing next-generation vaccines. It facilitates both the rapid design and evaluation of new and emerging vaccines and identifies individuals unlikely to be protected by vaccine. We describe a general-purpose machine-learning framework, DAMIP, for discovering gene signatures that can predict vaccine immunity and efficacy. DAMIP is a multiple-group, concurrent classifier that offers unique features not present in other models: a nonlinear data transformation to manage the curse of dimensionality and noise; a reserved-judgment region that handles fuzzy entities; and constraints on the allowed percentage of misclassifications. Using DAMIP, implemented results for yellow fever demonstrated that, for the first time, a vaccine's ability to immunize a patient could be successfully predicted (with accuracy of greater than 90 percent) within one week after vaccination. A gene identified by DAMIP, EIF2AK4, decrypted a seven-decade-old mystery of vaccination. Results for flu vaccine demonstrated DAMIP's applicability to both live-attenuated and inactivated vaccines. Results in a malaria study enabled targeted delivery to individual patients. Our project's methods and findings permit highlighting and probabilistically prioritizing hypothesis design to enhance biological discovery. Moreover, they guide the rapid development of better vaccines to fight emerging infections, and improve monitoring for poor responses in the elderly, infants, or others with weakened immune systems. In addition, the project's work should help with universal flu-vaccine design. © 2016 INFORMS.

Public health experts agree that the best strategy to contain a pandemic, where vaccination is the prophylactic treatment but vaccine supply is limited, is to give higher priority to higher-risk populations. We derive a mathematical decision framework to track the effectiveness of prioritized vaccination through the course of a pandemic. Our approach couples a disease-propagation model with a vaccine queueing model and an optimization engine to determine optimal prioritized coverage in a mixed-vaccination strategy. This demonstrably minimizes infection and mortality. Given estimated outbreak characteristics, vaccine inventory levels, and individual risk factors, the study reveals an optimal coverage for the high-risk group that results in the lowest overall attack and mortality rates. This knowledge is critical to public health policy makers for determining the best strategies for population protection. This becomes particularly important in determining when to switch from a prioritized strategy emphasizing high-risk groups to a nonprioritized strategy in which the vaccine becomes available publicly. Our analysis highlights the importance of uninterrupted vaccine supply. Although the 2009 H1N1 supply, received in interrupted batches, eventually covered over 30 percent of the population, the resulting attack and mortality rates are significantly inferior to those in a scenario where only 20 percent of the population is covered with an uninterrupted supply. We also learned that early vaccination is important. Contrasting the 2009 H1N1 supply to a 10 percent uninterrupted supply, a three-week delay in vaccination reduces the 9.9 percent infection reduction of the former to a mere 0.9 percent. The optimal trigger for switching from prioritized to nonprioritized vaccination is sensitive to infectivity and vulnerability of the high-risk groups. Our study further underscores the importance of throughput efficiency in dispensing and its effects on the overall attack and mortality rates. The more transmissible the virus is, the lower the threshold for switching to nonprioritized vaccination. Our model, which can be generalized, allows the incorporation of seasonality and virus mutation of the biological agents. The system empowers policy makers to make the right decisions at the appropriate time to save more lives, better utilize limited resources, and reduce the health-service burden during a pandemic event.