题 目: Gene relationship network modeling and prediction
报告人: Ye Yuan（袁野）, Ph.D.
Machine Learning Department, School of Computer Science at Carnegie Mellon University
时 间: 9月14日（周一）9:00-10:00
地 点: Online (Zoom会议)
会议 ID：640 8468 0249
主持人: Lucas Carey
In this report, I will present how to use model-driven and data-driven strategies to model and predict gene relationship network behavior and structure. Specifically, I will first introduce how we modeled and quantified a three-node ceRNA minimal network using differential equations and synthetic gene circuits. Secondly, to deal with large scale networks, we used deep learning to infer pairwise gene interaction and causality with an image-like joint PDF input and it was found that the two above approaches can get similar new discoveries. Thirdly, I will mention how to infer extracellular gene relationship with additional spatial information. Finally, I plan to discuss future work that how to combine the two strategies to extend network studies, especially on complex network behavior prediction and control.
Ye Yuan is a postdoc at Machine Learning Department, School of Computer Science at Carnegie Mellon University, advised by Prof. Ziv Bar-Joseph. He received his PhD from Automation Department at Tsinghua University in 2017, advised by Prof. Yanda Li and Xiaowo Wang. He received his bachelor degree from Automation Department at Xi’an Jiaotong University in 2012. He also worked as a senior R&D engineer at Baidu Big Data Lab (Beijing) in 2017.
His research lies in bioinformatics and machine learning, focusing on machine learning and physical model with applications to relationship inference in genomics.