A Robust Semiparametric Model for the Analysis of Human Microbiome

• Activity Number: 180 - Statistical Methods in Phylogenetics and Metagenomics
• Type: Contributed
• Date/Time: Tuesday, August 4, 2020 : 10:00 AM to 2:00 PM
• Sponsor: Section on Statistics in Genomics and Genetics
• Author(s): Jinyuan Liu* and Xin Tu
• Companies: University of California, San Diego and University of California, San Diego
• Keywords: Between-subject attributes; Copula; Microbiome; Robust inference; U-statistics based generalized estimating equations

Abstract

The human microbiome plays an important role in disease and health and identifying factors associated with microbiome composition provides insights on the inherent disease mechanisms. By amplifying and sequencing the 16S rRNA gene, with highly similar sequences binned together, we obtain Operational Taxonomic Units (OTU) abundances for each subject. Due to the high dimensionality and non-normality of the OTU abundances, distance-based measures are often applied and one popular measure is the beta-diversity between individuals. Analyses of such between-subject attributes are not amenable to the predominant within-subject based statistical paradigm, such as t tests and linear regression. We propose a new approach to model the beta-diversity as a response within the regression setting by utilizing the functional response models (FRM), a class of semi- parametric model for between- as well as within-subject attributes. The new approach not only addresses limitations of current methods for cross-sectional study data, but also pro- vides a basis for longitudinal and other clustered data in the future. The proposed approach is illustrated with both real and simulated data.