报告人：Jie Peng
报告地点：研究生军事多媒体教室
报告时间：2015年7月12日13:30-14:20
主讲人简介：
加州大学戴维斯分校UC Davis
Contemporary large-scale omic studies often collect many types of high throughput data on each subject, including copy number alterations, RNA expressions and protein expressions. This provides us an unprecedented opportunity to study the biological
system as a whole, from DNA to RNA and to protein. One question of great interest is how different biological components interact with each other. In statistics, interactions are often modeled by conditional dependencies. This makes graphical models perfect tools for studying genetic regulatory relationships.
In this talk, we will give a brief introduction of graphical models and various algorithms for fitting graphical models. We will focus on structure learning algorithms in the high-dimensional regime. Applications to build genetic regulatory networks (GRNs) using high throughput omic data will be discussed throughout the talk.
