A0942
Title: Beyond the one-size-fits-all: A deep learning method to identify subgroup-specific biomarkers
Authors: Anja Logan - University of Minnesota (United States) [presenting]
Sandra Safo - University of Minnesota (United States)
Abstract: The heterogeneity in many complex diseases has spurred efforts to leverage multiomics and phenotypic data to identify biomarkers of disease risk and progression to better understand the underlying physiology. These attempts focus mainly on the general population, use few molecular factors, hardly account for social determinants of health (SDoH), and establish simple associations, limiting the ability to better characterize health for disadvantaged populations. A broader, systems-level perspective centered on the totality of SDoH, multiomics, and phenotypic data is proposed, using innovative interpretable deep learning (DL) methods to better understand and help address health disparities. The proposed DL method jointly integrates data from multiple sources and predicts a clinical outcome while yielding common and subgroup-specific variable selection. Simulations are used to demonstrate the effectiveness of the proposed and other methods in the literature. Real data analyses would be conducted.
