(364w) Modeling Affinity Maturation to Recapitulate Germinal Center Dynamics

Through a combination of methods ranging from atomistic molecular dynamics to deep reinforcement learning, my PhD work has been focused on answering the how and why to questions related to the immune response. In my next steps, I aim to leverage my diverse background in computational biology, deep learning, and viral immunology to aid in the design of better vaccines and therapeutics. While some of my past work has provided insights into the fundamentals of how the adaptive immune response functions, my engineering background has always tied these findings back to how they might inform the design and production of potential interventions. My research interests will certainly evolve over time—as they have throughout graduate school—but the underlying motivation remains the same: How can we alleviate the burdens created by disease?

By cultivating a highly interdisciplinary environment around me in graduate school, I have had the privilege of engaging with a multitude of disciplines enabling me to speak the same language as a computer scientist, immunologist, and engineer. This background has prepared me to identify and communicate how computational models can inform experimental or clinical results. As I approach graduation and beyond, I hope to continue to use and develop my expertise in both modeling and communication with interdisciplinary teams. My research interests will continue to evolve, just as the projects we work on evolve, but I am confident my training has prepared me to adapt to the ever-changing technologies the biotech sector has to offer.

Research Abstract:

Highly mutable infectious disease pathogens such as HIV and influenza evolve too rapidly for the human immune system to contain them, allowing them to circumvent traditional vaccination approaches, and cause over one million deaths annually. To combat these pathogens and others, we must develop tools to aid in the design of robust, potent, and universal immunogens. Agent-based models can be used to simulate the complex interactions and dynamics that occur in the lymph nodes between immune cells and pathogen-like proteins (antigens) during affinity maturation—the process by which antibodies evolve. As such, compared to existing experimental approaches, agent-based models offer a safe, low-cost, and rapid route to study the immune response to vaccines spanning a wide range of design variables. Previously, models of affinity maturation have been limited in their ability to leverage the sequence and structure of the antibody/antigen complex, replicate the immunological dynamics found in vivo, or both. Herein, we build upon our recent efforts incorporating antibody nucleotide sequences, antigen amino acid sequences, and 3D crystal structures. Through the development of both more physiologically relevant B cell dynamics and an explicit antigen capture model, we accurately recapitulate a variety of relevant metrics seen in vivo. Importantly, the model obtains these results without sacrificing the sequence or atomistic-level details of the evolving B cells. Preserving these important biological factors allows for direct examination of the resulting antibody sequences for clinically relevant clonotypes observed in the general population following exposure to antigens of interest via infection or vaccination. By accurately capturing both the dynamics of affinity maturation and the details of the resulting immune response, we believe this model will enable the rational design and preliminary screening of immunogen candidates for a variety of pathogens which have thus far proven too difficult or too expensive to test by traditional means.