当前位置: 首页 >> 学术活动 >> 正文

学术报告:Toward More Sensitive Differential Expression Analysis on RNA-Seq Data

发布人:日期:2017-11-28浏览数:

报告题目:Toward More Sensitive Differential Expression Analysis on RNA-Seq Data

报告人: Tao Jiang教授,Department of Computer Science and Engineering

University of California, Riverside, and

School of Information Science and Technology, Tsinghua University

报告时间:2017年11月30日10:00

报告地点:量子楼410报告厅

报告摘要:As a fundamental tool for discovering genes involved in a disease or biological process, differential gene expression analysis plays an important role in genomics research. High throughput sequencing technologies such as RNA-Seq are increasingly being used for differential gene expression analysis that was dominated by the microarray technology in the past decade. However, inferring differentially expressed genes based on the observed difference of RNA-Seq read counts has unique challenges that were not present in microarray-based analysis. An RNA-Seq based differential expression analysis may be biased against genes with low read counts since the difference between genes with high read counts is more easily detected. Moreover, analyses that do not take into account alternative splicing often miss genes that have differentially expressed transcripts. In this talk, we introduce two novel methods for enhancing differential expression analysis. One uses a markov random field (MRF) model to integrate RNA-Seq data with coexpression data and the other represents independent alternative splicing events by decomposing the splice graph of a gene into special modules (called alternative splicing modules or ASMs). Our extensive experiments on simulated data and real data with qPCR validation demonstrate that these enhancements lead to more sensitive differential expression analyses and better classification of cancer subtypes, cell types and cell-cycle phases.

报告人简介:美国加州大学河边分校计算机系教授,清华大学国家信息科技实验室千人讲席教授,计算机科学协会ACM(Association for Computing Machinery ) 和 美国科学促进协会 (American Association for the Advancement of Science)的会士(Fellow)。姜涛教授的研究方向包括:组合算法的设计与分析,生物学信息学,计算复杂度,信息收集及提取的计算方法等。他在主流的计算机理论和生物信息学杂志发表了260多篇文章。是五十多个国际会议的委员会委员和主席,六个学术基金评审委员会委员。同时,他(曾)是组合优化杂志 (Journal of Combinatorial Optimization),计算机科学与技术杂志 (Journal of Computer Science and Technology),生物信息与计算生物学杂志 (Journal of Bioinformatics and Computational Biology),BMC 生物信息杂志(BMC Bioinformatics),IEEE/ACM计算生物学及生物信息杂志(IEEE/ACM Transactions on Computational Biology and Bioinformatics),算法杂志(Algorithmica),计算机与系统科学杂志(Journal of Computer and System Sciences)编委。