学术预告

Understanding complex systems by model reduction using differential geometry

发布日期:2017-08-29 发表者: 浏览次数:

报告题目:Understanding complex systems by model reduction using differential geometry

报告人: 黄磊 博士

报告时间:2017年9月1日(周五)10:00

报告地点:逸夫楼C座314会议室

摘要:

Complex systems, including cells, crops, neural networks and power systems, are notoriously hard to study and are the frontiers of science. A recently developed method, called Manifold Boundary Approximation Method (MBAM), promises to help the studies of complex systems. Given a mathematical model of a complex system, MBAM simplifies the model by reducing its degrees of freedom as measured by the number of parameters, while maintaining the mechanistic relations with the original model and hence the interpretability of the simplified models. The resulting simplified models better reveal how system behaviors emerge from microscopic interactions than the original model, a central goal in the studies of complex systems. As the name suggests, MBAM is geometric in nature (it views a model as a manifold) and is hence general and automatic. Application of MBAM is illustrated mainly using two models from systems biology: nerve growth factor signaling network and photosynthesis metabolic network. Some other general insights and useful results on mathematical modeling as derived from the geometric framework are also discussed.

报告人简介:

黄磊博士于2007年在中山大学获得生物学及统计学学士学位,2017于美国康奈尔大学获得计算生物学理学博士学位,目前在Craig Venter创办的Human Longevity公司任职。主要研究方向为生物系统的数学建模及生物数据挖掘,在Cell Metabolism, PNAS, eLife, PLOS Computational Biology等国际著名期刊发表多篇论文。