High resolution profiling and computational modeling of influenza virus-immune dynamics during natural infection - Abstract Seasonal influenza viruses (IVs) cause hundreds of thousands of deaths every year, despite widespread pre- exposure and vaccination. To design more effective vaccines against both seasonal and pandemic IVs, there is an urgent need to better understand how the complex interplay between viral and immune dynamics in humans influence infection outcomes and transmission risk. Several critical fundamental knowledge gaps hobble the development and evaluation of next-generation IV vaccines. One is that we have a very poor understanding of which specific features of the enormously complex human immune response to IV can serve as reliable correlates of protection from severe infection outcomes and forward transmission risk. This makes it difficult to predict the performance of vaccine candidates in the human population. Similarly, we do not understand how infection or vaccination drive the evolution of immune memory repertoires across highly heterogeneous individuals. This makes it hard to design/evaluate novel antigens that reliably reshape the immune memory repertoire to effectively protect against future strains. To address these critical gaps in our understanding of functional IV immunity in humans, we will build upon a unique clinical study already underway on the University of Illinois campus that uses a frequent PCR screening program to identify individuals during the first few days of IV infection. By combining repeated blood draws with daily virus sampling following infection, we can generate both an unprecedented high-resolution longitudinal profile of viral shedding dynamics and a rich, highly multidimensional profile of immune status before, during, and after infection. Our multidisciplinary team of experimentalists and computational modelers will fit the virological and immunological data we collect to mechanistic models of virus-immune dynamics to infer key viral and immune dynamics features from each infected individual. We will then apply novel statistical and machine learning approaches to achieve two primary aims: (1) identify reliable immune correlates of reduced viral shedding and transmission risk, and (2) define how both viral and host factors shape the evolution of the anti-IV B cell repertoire and functional antibody landscape. Together, these aims will identify new host and viral features that correlate with desirable, protective memory responses against IV that can be incorporated into the design and evaluation of next generation, universal IV vaccines.
