Conference Program Home
  My Program

All Times EDT

Abstract Details

Activity Number: 175 - Computational Methods for Complex Data Challenges
Type: Invited
Date/Time: Monday, August 8, 2022 : 2:00 PM to 3:50 PM
Sponsor: Section on Statistical Computing
Abstract #320645
Title: Joint Matrix Decomposition Regression for Supervised Multi-Omics Data Integration
Author(s): Yue Wang* and Tim Randolph and Jing Ma and Ali Shojaie
Companies: Arizona State University and Fred Hutch Cancer Center and Fred Hutch Cancer Center and University of Washington
Keywords: joint matrix decomposition; high-dimensional data; data integration; supervised learning; human microbiome

Diagnosis and treatment of human diseases require joint interpretation of molecular variations at multiple levels. For example, while the human microbiome has been proved associated with host health, limited knowledge is known about the underlying mechanism of these associations. One potential mechanism is microbial metabolism, which may affect host metabolic processes. Thus, an integrative analysis of microbiome, metabolome, and host health may shed light on the biological mechanism of microbiome-host interactions. However, this analysis is challenging because it involves multiple high-dimensional data sets and complex correlations between microbes, between metabolites, and between a microbe and a metabolite. We propose a high-dimensional data integration tool based on the joint matrix decomposition (called JMDR) to examine joint associations between multi-omics data and an outcome of interest. The JMDR is able to examine the global joint effects on the outcome as well as the contributions each biomarker makes to the joint effects. Thus, the JMDR facilitates the understanding of the interplay between the multi-omics variations and human health.

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

Back to the full JSM 2022 program