Phylogeny-Guided Microbiome OTU-Specific Association Test (POST)

• Activity Number: 197 - SPAAC Poster Competition
• Type: Topic Contributed
• Date/Time: Monday, August 8, 2022 : 2:00 PM to 3:50 PM
• Sponsor: Section on Statistics in Genomics and Genetics
• Author(s): Caizhi Huang* and Benjamin Callahan and Michael C. Wu and Shannon T. Holloway and Hayden Brochu and Wenbin Lu and Xinxia Peng and Jung-Ying Tzeng
• Companies: Bioinformatics Research Center, North Carolina State University and North Carolina State University and Fred Hutchinson Cancer Research Center and North Carolina State University and Bioinformatics Research Center, North Carolina State University and North Carolina State University and North Carolina State University and North Carolina State University
• Keywords: Association test; Phylogenetic tree; Kernel machine regression; Operational taxonomic unit; Microbiome

Abstract

The link between host conditions and microbiome profiles contains important information about the microbial role in human health. Traditional association testing frameworks are challenged by the high-dimensionality and sparsity of microbiome data. Incorporating phylogenetic information is used to address these challenges with the assumption that evolutionarily similar taxa tend to behave similarly. However, this assumption may not always be valid and phylogenetic information should be incorporated in a data-supervised fashion. We propose a local collapsing test called Phylogeny-guided microbiome OUT-Specific association Test (POST). In POST, how much information to borrow from the neighboring OTUs is supervised by phylogenetic distance and outcome-OTU association. POST is constructed under kernel machine framework to accommodate complex OTU effects and extends kernel machine microbiome tests from community-level to OTU-level. Using simulation, we showed that when the phylogenetic tree is informative, POST has better performance than existing OTU-level association tests. We show that POST can identify more outcome-associated OTUs in real data application on bacterial vaginosis.