Online Program Home
My Program

Abstract Details

Activity Number: 234 - Novel Statistical Methods for High-Dimensional Microbiome and Metagenomics Data Analysis
Type: Topic Contributed
Date/Time: Monday, July 29, 2019 : 2:00 PM to 3:50 PM
Sponsor: Section on Statistics in Epidemiology
Abstract #301743 Presentation
Title: Estimating and Testing the Microbial Causal Mediation Effect with High-Dimensional and Compositional Microbiome Data
Author(s): Chan Wang* and Jiyuan Hu and Martin Blaser and Huilin Li
Companies: Division of Biostatistics, NYU School of Medicine and New York Unversity School of Medicine and New York University School of Medicine and Rutgers University and NYU School of Medicine
Keywords: Causal mediation analysis; Composition; Dirichlet regression; High-dimension; Hypothesis testing; Microbiome
Abstract:

Microbiome associating findings encourage scientists to dive deeper to uncover the causal role of microbiome in the underlying biological mechanism, and have led to applying statistical models to quantify causal microbiome effects and to identify the specific microbial agents. However, there are no existing causal mediation methods specifically designed to handle high dimensional and compositional microbiome data. In this manuscript, we propose a rigorous Sparse Microbial Causal Mediation Model (SparseMCMM), consisting of linear log-contrast regression and Dirichlet regression, to estimate the causal direct effect of treatment and the causal mediation effects of microbiome at both the community and individual taxon levels, in the typical three-factor (treatment, microbiome and outcome) causal study design. Two hypothesis tests on the overall mediation effect are proposed as well. Extensive simulated scenarios and real data application show that SparseMCMM has excellent performance in estimation and hypothesis testing. With SparseMCMM, we can get a clear and sensible causal path among treatment, high-dimensional and compositional microbiome and outcome.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2019 program