鱼虾蟹游戏

报 告 题 目：Machine Learning-Based Approached for Integrating Single Cell and Clinical Data for Disease Marker Discovery

主 讲 人：黄 昆

单 位：印第安纳大学

时 间：11月16日10:00

腾 讯 ID: 583-9798-6772

摘 要：

For complex diseases such as cancers and Alzheimer’s diseases, it is of great translational value to identify molecular markers associated with diseases that can facilitate diagnosis, prognosis, and prediction of treatment responses. However, these diseases are usually high heterogeneous and the disease tissues usually constitute multiple types of cells and thus it is of interest to identify subpopulations of cells that are associated with diseases or disease outcomes. Such disease associated cells and their markers are often difficult to identify from bulk tissue genomic data. With the progresses in technologies such as scRNA-seq enabled researchers to study molecular profiles of the cells at single cell level. An interesting and important problem is then how to link the scRNA-seq data with both bulk tissue gene expression and clinical data to identify the disease associated subpopulations of the cells and their markers. In this talk I will introduce the background of the problem and the applications of machine learning methods in addressing these issues. Specifically, I will present our recent work on developing a transfer learning framework for integrating scRNA-seq, bulk tissue RNA-seq, and clinical data with applications in cancer and Alzheimer’s diseases.

简 介：

黄昆, 1996年毕业于清华大学获生物学理学与电子计算机工学双学士学位, 后于美国伊利诺伊大学香槟校区(UIUC)获得生理学、电子工程和数学等三个硕士学位，2004年获得电子与计算机工程学博士学位。2010年获俄亥俄州立大学终身教职。2017年加入印第安纳大学医鱼虾蟹游戏 参与领导精准健康计划。现担任印第安纳大学医鱼虾蟹游戏 与Fairbanks公共卫生鱼虾蟹游戏 生物统计与健康数据科学系主任，印第安纳大学精准健康数据科学与信息学主任，基因组数据科学讲席教授，印第安纳大学Simon综合癌症中心副主任。2018年当选美国医学与生物工程鱼虾蟹游戏 (AIMBE)会士。主要研究方向包括生物信息学，医学图像分析，医疗大数据，机器学习及其在癌症研究及神经科学等方面的应用，在《Nature》、《Science》、《Cell》等杂志和会议上发表研究论文200余篇。