“数理论坛”第77期：Normalization methods for metagenomics compositional data

报告时间

2019年1月4日下午14:00——16:00

报告地点

东教学楼1304

报告人

方志德教授

报告人

简介

方志德博士：美国路易斯安那州立大学新奥尔良卫生科学中心,   公共卫生学院生物统计学系教授,   系主任；医学院遗传学系教授。他是美国国家卫生研究院资助的路易斯安那临床和转化中心的统计专家, 生物信息/生物计算联络员。方教授的研究领域包括: 构建优化实验设计, 可靠性研究, 生存分析, 规范矩理论。方教授同时致力于开发有效的、应用于生物医药、   生物信息研究的统计方法。这些应用包括神经科学、癌症基因组学、全基因组的基因表达、微RNA、通由路分析、DNA拷贝数变异和宏基因组学等。

报告摘要

In recent years, metagenomics, as a combination of research techniques   without the process of cultivation, has become more and more popular in   studying the genomic/genetic variation of microbes in environmental or   clinical samples. Though generated from similar sequencing technologies,   there is increasing evidence that metagenomic sequence data may not be   treated as another variant of RNA-Seq.count data, especially due to its   compositional characteristics. While it is often of primary interest to   compare taxonomic or functional profiles of microbial communities between   conditions, normalization for library size is usually an inevitable step   prior to a typical differential abundance analysis. Some methods have been   proposed for such normalization. But the existing performance evaluation of   normalization methods for metagenomic sequence data does not adequately   consider the compositional characteristics. In this paper, we assess the   normalization methods in the literature and study the impact of the methods   on differential abundance analysis, using simulated and real life data.

邀请人

肖海军教授

2018年12月20日