Pepper Yitong Huang: Novel mathematical tools for determining physiological time

Humans have an internal timekeeping system, with every cell possessing a daily (circadian) clock that regulates sleep, performance, and virtually every physiological function. These self-sustained oscillators not only synchronize with one another but also adapt to environmental cues. In the first half of this talk, I introduce a mathematical model of coupled oscillators. Using a mathematical ansatz, specifically the contraction to low-dimensional manifolds, to simplify coupled oscillator problems, I demonstrate how this model effectively captures the dynamics of circadian clocks. With the explanatory power of this bottom-up mechanistic modeling approach, I then show how one can optimize schedules to maximize productivity and minimize jet lag. In the second half, I discuss how high-dimensional time-series data from high-throughput omics assays offer new opportunities while also posing challenges in characterizing the dynamics of circadian rhythms. I present TimeMachine, a novel machine-learning algorithm designed to estimate phase from high-dimensional, noisy data, such as gene expression in human blood. Finally, I will discuss my ongoing work on optimal timing strategies for temozolomide treatment in glioblastoma. This talk brings together mathematics, statistics, biology, and technology, presenting a new paradigm in utilizing mathematical principles to drive scientific discoveries.

Speaker's webpage: https://sites.google.com/view/pepperhuang/bio