A multi-source data integration approach reveals novel associations between metabolites and renal outcomes in the German Chronic Kidney Disease study

Abstract

Omics data facilitate the gain of novel insights into the pathophysiology of diseases and, consequently, their diagnosis, treatment, and prevention. To this end, omics data are integrated with other data types, e.g., clinical, phenotypic, and demographic parameters of categorical or continuous nature. We exemplify this data integration issue for a chronic kidney disease (CKD) study, comprising complex clinical, demographic, and one-dimensional 1H nuclear magnetic resonance metabolic variables. Routine analysis screens for associations of single metabolic features with clinical parameters while accounting for confounders typically chosen by expert knowledge. This knowledge can be incomplete or unavailable. We introduce a framework for data integration that intrinsically adjusts for confounding variables. We give its mathematical and algorithmic foundation, provide a state-of-the-art implementation, and evaluate its performance by sanity checks and predictive performance assessment on independent test data. Particularly, we show that discovered associations remain significant after variable adjustment based on expert knowledge. In contrast, we illustrate that associations discovered in routine univariate screening approaches can be biased by incorrect or incomplete expert knowledge. Our data integration approach reveals important associations between CKD comorbidities and metabolites, including novel associations of the plasma metabolite trimethylamine-N-oxide with cardiac arrhythmia and infarction in CKD stage 3 patients.

Integrated Analysis of Gut Microbiome and Lipid Metabolism in Mice Infected with Carbapenem-Resistant Enterobacteriaceae

“…Systematic integration of multiple layers of omics datasets has been indicated to obtain more reliable results and reduce the false-positive risk [ 26 , 27 ], which can provide a basis for generating testable hypotheses and gaining mechanistic insights into the pathophysiology of multiple complex diseases in post-integration analyses [ 28 , 29 ]. In this study, the multi-omics technologies, including metagenomics, metabolomics and lipidomics, comprehensively characterized the alterations of gut microbiome and lipid metabolism occurring during the early infection of CRE in mice.…”

Educational Attainment Is Associated With Kidney and Cardiovascular Outcomes in the German CKD (GCKD) Cohort

“…As some rare causes of CKD etiology were small in sample size, we did not adjust further the regression analysis. Mixed graphical models (MGMs) 19 , 20 , 21 ( Supplementary Methods ) were used to identify baseline variables independently associated with educational attainment. An MGM is an unbiased and data-driven powerful tool to eliminate indirect associations discovered by routine univariate screening approaches and can reveal associations adjusted for all other variables in the data set.…”

“…An MGM is an unbiased and data-driven powerful tool to eliminate indirect associations discovered by routine univariate screening approaches and can reveal associations adjusted for all other variables in the data set. 19 , 20 , 21 The identified variables were considered as mediator candidates between educational attainment and outcomes and were used in the Cox proportional hazard regression models (see model 2). A mediator is the variable that causes mediation between the exposure and the outcome variable (see subsequent discussion).…”

Multi-Omics Integrative Analysis Coupled to Control Theory and Computational Simulation of a Genome-Scale metabolic Model Reveal Controlling Biological Switches in Human Astrocytes Under Palmitic Acid-Induced Lipotoxicity

“…Dimensionality reduction is a mathematical way to reduce the complexity of a data set while increasing the statistical power of analysis by reducing the burden of multiple tests (Zierer et al, 2016;Altenbuchinger et al, 2019;Wörheide et al, 2021), we accomplished this reduction by means of principal component analysis (PCA). In our case, PCA was applied to the transcriptome and proteome data sets, transforming the unique omic variables into a lower-dimensional subspace that maximizes the retention of variance within the data by finding orthogonal linear combinations of the original variables, saving as much information as possible.…”