讲座题目：高维成分数据的若干统计推断理论

主办单位：三峡数学研究中心/理学院

报告专家：李正帮

报告时间：2023年7月6日（周四）下午19:00

报告地点：腾讯会议（ID：900-122-320）

专家简介：李正帮博士，华中师范大学数学与统计学学院副教授、硕士生导师，武汉大学数学与统计学院硕士、博士，美国宾西法利亚大学生物统计系博士后，主持或参与国家自然科学基金多项，在《Scientific Report》等期刊上发表重要学术论文20多篇，主要研究兴趣：大数据分析与统计建模，生物统计，统计遗传学，全基因组关联分析，全表现型关联分析，复杂疾病统计分析等。

报告摘要：we propose a novel test for comparing two high-dimensional microbiome abundance data matrices based on the centered log-ratio transformation of the microbiome compositions. The test p-value has a closed-form solution from the derived asymptotic null distribution. We also study the asymptotic statistical power against sparse alternatives that are common in microbiome studies. The proposed test is maximum-type equal-covariance-assumption-free (MECAF), which makes it suitable for studies that compare microbiome compositions between conditions. Our simulation studies show that the proposed MECAF test has higher power than competing methods while controlling the type I error rate well under various scenarios. We further demonstrate the usefulness of the proposed test with two real microbiome data analyses. The source code of the proposed method is freely available at https://github.com/Jiyuan-NYU-Langone/MECAF. MECAF is a flexible and efficient differential abundance test for analyzing high-throughput microbiome data. The proposed new method will help us to discover shifts in microbiome abundances between disease and treatment conditions, enhancing our understanding of the disease and ultimately improving clinical diagnosis and treatment.