学 术 讲 座
主题：Reference-free Learning with Multiple
Metagenomic Samples
主讲人：Wenxuan Zhong （美国佐治亚大学统计系教授）
时间：2017年07月05日 （周三）上午 10:00-11:30
地点：888.3net新浦京游戏津南校区计算机与控制工程学院（信息东楼）523会议室
Abstract： The major goals of metagenomics are to identify and study the entire collection
of microbial species in a set of targeted samples through sequencing bulk DNA
extracted directly from the samples. So far, however, there has not been an
effective reference-free tool for dissecting the complex information revealed
by the sequencing. In this talk, I will present a statistically based algorithm
to simultaneously identify microbial species and estimate their abundances in
multiple metagenomic samples without using any reference genome. Compared with
existing reference-free methods that are mostly based on k-mer distributions,
this new approach can achieve a higher species identification accuracy, and is
particularly powerful when the sequencing coverage is low. We demonstrate the
performances of the new method through both simulation and real metagenomic
studies.
Biography：钟文瑄，现任美国佐治亚大学统计系副教授(Department
of Statistics, University of Georgia)，大数据分析研究室主任，致力于发展降维方法以及这些方法在机器学习、生物信息等学科问题中的应用，近年来主要研究发展大数据降维的算法及其理论基础。钟文瑄教授本科毕业于888.3net新浦京游戏数学系，获美国普渡大学统计学博士，哈佛大学博士后。2007年8月起，任职于美国伊利诺伊大学香槟分校(University of Illinois at
Urbana-Champaign)助理教授；2014年被美国佐治亚大学引进. 钟文瑄教授在《Journal of Royal Statistical
Society Ser B》，《Journal of American
Statistical Association》等国际顶尖期刊上发表多篇论文，形成了极高的影响力。钟文瑄教授的研究被英特尔（Intel）公司，匹兹堡超级计算中心评价为大数据分析的突破性工作；钟文瑄教授担任学术期刊Statistica Sinica等的副主编。
