Inferring multi-scale dynamics underlying behavior in aging C. elegans

点击翻译按钮获取中文摘要

Project Summary Physiological processes such as aging must arise from activities and events spanning multiple time scales. Although important, the dynamics of these processes are difficult to measure and study. In multicellular model organisms for aging, the freely living nematode C. elegans is among the best studied, and yet, most of the aging studies are limited to simple outcomes such as lifespan and completely ignore the process through which aging occurs. This limitation is in part due to the lack of economical and scalable technologies to acquire detailed data during the aging process (recording behavior and neural dynamics on individual basis using conventional approaches are very expensive); further, there has not been a well-established theoretical (and computational) approach to model the aging behavior and make connections between short- and long-term dynamics. The lack of these tools result in the very limited description and understanding of the mechanisms of healthspan, even in excellent genetic model systems as C. elegans. The goal of this application is to define behavioral states and dynamics in the aging process and examine the neural origins of the behavioral dynamics. First, high-throughput experimental systems and robust computational pipelines will be engineered. The tools will then be used to characterize short- and long-term behavior dynamics in aging, especially in response to food availability that results in modulations of longevity and health. Models will be built to connect the neural dynamics to behavioral dynamics, and the models will address whether it is modulations in neural dynamics that lead to changes in behavioral states and different long-term physiological outcomes. The proposed project is innovative, because it is the first time multi-scale behavioral dynamics is recorded in large number of individuals and fully characterized using stochastic dynamical system models in aging process; it is also the first time that behavioral dynamical models are connected to and explained by neural dynamical models. The proposed work is significant, because of both the tools and insights it generates. Scientifically, the ability to define internal states and understand the aging trajectory, especially with insights to the neural origin of the observed dynamics under a variety of longevity-inducing conditions, will point to potential strategies to influence healthy aging. Technologically, we envision that both the high-throughput behavioral recording platform and the neural imaging pipeline will be useful beyond C. elegans, and that the theoretical and computational pipelines can potentially be generalized to many other experimental systems, including mice, non-human primates, and humans. The rich data set for aging behavior will also benefit other aging researchers.