Math Biology Seminar: Pavitra Roychoudhury

Individual-based models are widely used to describe the characteristics and dynamics of single organisms or particles in a population. A useful feature of these models is the ability to easily incorporate stochasticity and spatial structure at the individual level and, as a result, these models are well-suited to addressing questions in evolutionary biology and infectious diseases. In this talk, I will describe two models we have developed that incorporate these features. In the first project, we developed a stochastic, mechanistic model to predict the effectiveness and toxicity of therapies currently being developed to cause targeted disruption of latent viral genomes for the cure of diseases like HIV. We fit our model to flow cytometry data from multiple experiments aimed at optimizing engineered DNA cleavage enzymes delivered to cells using adeno-associated viruses (AAV) vectors. The model predicts the number of transgenes delivered, the level of expression and amount of cytotoxicity produced as a function of dosage for a given AAV serotype. We then use the model to predict the therapeutic index for a candidate therapeutic molecule and determine the optimal dosage, serotype and promoter for delivering the molecule to infected cells. In the second project, we developed a spatially structured, individual-based model of HSV spread in epithelial tissue with the goal of understanding how tissue-resident memory CD8+ T-cells (TRMs) control a reactivating HSV infection. CD8+ TRMs are a relatively recent discovery and characterizing their role and interactions with other T-cell compartments is vital for designing effective vaccine strategies against HSV. Our model incorporates mechanisms like viral diffusion, cell-cell spread, patrolling by TRMs, trafficking of effector memory T-cells (TEMs) from lymph nodes and effector functions of CD8s including tissue-wide alarm functions.