To defend against rapidly evolving pathogens, jawed vertebrates have specialized cells—lymphocytes—that
evolve during an individual's lifetime to mount adaptive immune responses. Massive somatic diversity is
maintained in loci that encode receptors on lymphocytes that can detect foreign antigen. B cells—lymphocytes
that make antibodies—bind antigen with the B cell receptor (BCR), and diversify in microanatomical structures
called germinal centers (GCs) where they proliferate while mutating the BCR. This process of af¿nity matura-
tion is classically understood to impose selection for increased antigen binding af¿nity: surviving cells mature
to become high-af¿nity memory B cells or plasma cells that secrete antibodies (the soluble form of the BCR).
There is great potential for productive dialog between experiment, theory, and computation to learn new
immunology and new evolutionary dynamics from these systems. Indeed, recent studies reveal a literature
con¿icted regarding GC evolution as adaptive toward high af¿nity. The hypothesis of this application is that GC
evolutionary dynamics balance adaptation towards antigen speci¿city with neutral diversi¿cation that forti¿es against
The research strategy is to develop quantitative models of GC evolution in a tight feedback loop with
model organism experimental design of increasing complexity. Using mouse models, control can be exerted
over B cell repertoire diversity, receptor af¿nity, and antigen targeting. Evolution in GCs can be tracked using
both receptor sequencing and lineage tracing technology. Aim 1 will analyze GC BCR evolution in a mouse
model with a monoclonal BCR and a model antigen exposure, such that all GC reactions constitute replicated
evolution from the same ancestral state. Aim 2 will develop theoretical and computational tools to infer ¿tness,
convergence, and contingency from the (quasi-)replicated evolutionary dynamics that characterize both our
experimental models and natural repertoires. Aim 3 will design and analyze two mouse models of increasing
repertoire complexity, and investigate recall responses to modi¿ed antigens.
The training plan is designed to synthesize expertise in theoretical and computational evolutionary biology
with immunology, and advance both hard and soft skills necessary for a future as a independent investigator.
Co-sponsors Dr. Frederick Matsen and Dr. Kelley Harris combine expertise in mathematical and computa-
tional biology and immunology, and population genetics. B cell immunologist Dr. Gabriel Victora will be a
close collaborator; his lab will lead experimental development of the mouse models. I will also collaborate
with theoretical physicist Dr. Armita Nourmohammad, who has expertise in rapid evolutionary dynamics.
My thesis committee adds Dr. Phil Green and Dr. Joe Felsenstein as computational and theoretical resources.
This interdisciplinary research team is matched to the proposed aims as a synthesis of distinct ¿elds.