UCSF

Genetics is currently undergoing a ‘big-data’ revolution, where the advent of deep-sequencing has enabled researchers to routinely create massive datasets, yet the statistical analyses of these data remain challenging. Deep-sequencing has proved particularly useful in the field of evolution, where researchers can sequence rapidly adapting populations to observe evolution taking place in real-time. In this dissertation, I design and implement statistical tools to answer relatively simple evolutionary questions about complex, large, and rapidly adapting populations. Specifically, I develop methods to infer convergence in the antibodies of in vivo human B cells, selection in in vitro viral populations, and coevolution between in vivo populations of human antibodies and autologous human immunodeficiency virus (HIV). I show that the antibodies of human B cells will converge towards similar genetic sequence characteristics when presented with identical influenza vaccines, and that HIV and antibodies can exhibit genetic signatures of coevolution, although this is quite rare. I also show that genetic drift plays a significant role in an experimental population of HIV, and that this needs to be taken into account when inferring selection. Together, I hope that the analyses described in this work prove useful to other investigators with similar evolutionary questions.