Molecular Characterizations of the Pathophysiology of Type 2 Diabetes in Hispanic/Latino Populations

ABSTRACT Type 2 diabetes (T2D) is a common multifactorial metabolic condition characterized by persistently elevated blood glucose concentration. T2D is a major driver of global health burdens, with a disproportionate impact on historically marginalized populations, specifically, US Hispanic or Latino (HL) populations. HL populations have an increased prevalence of T2D compared to non-Hispanic whites; however, they are often underrepresented in genetic studies and have strikingly limited representation in multi-omic studies. The current work proposes to use deeply phenotyped cohorts in conjunction with multi-omic data at multiple timepoints to better understand the biological mechanisms contributing to T2D. Aim 1 will identify molecular markers associated with T2D cross-sectionally by leveraging transcriptomics, proteomics, and metabolomics data in multiple cohorts. We will perform differential abundance analyses in the available omic platforms from Mexican American participants in the Cameron County Hispanic Cohort (CCHC) and the Hispanic Community Health Study/Study of Latinos (HCHS/SOL) using linear regression. We will assess replication and generalizability of significant markers from each omics platform by analyzing Mexican American and non-Mexican American holdout sets from the CCHC and the HCHS/SOL. To explore molecular profiles associated with disease onset, Aim 2 will utilize longitudinal data across available omics to pinpoint signals associated with incident T2D. Using multiple time points collected from Mexican American participants in the CCHC and HCHS/SOL, we will perform multiple linear mixed model analyses. We will assess replication and generalizability of significant markers from the transcriptomics and metabolomics results in holdout sets from the CCHC and HCHS/SOL, as in Aim 1. Aim 3 will identify likely functional mechanisms underlying T2D risk and associated causal relationships. We will utilize ancestry-informed colocalization, Mendelian randomization, and phenome-wide association studies, to better characterize biological processes involved in T2D in HL individuals by leveraging electronic health record data available from BioVU and AllofUS. These analyses will explore causal and functional evidence for molecular features identified in Aims 1 and 2, our preliminary data, and the literature. We anticipate that results from Aim 3 will illuminate specific biological mechanisms contributing to risk and progression of T2D via genetic mechanisms. Together, these aims will characterize molecular factors associated with T2D and development in HL populations, enhancing our understanding of cellular changes associated with T2D in blood, and potentially illuminating clinical and therapeutic targets for future research.