Metabolomics and proteomics, the systematic analysis of small molecules and proteins, respectively, have been increasingly utilized in kidney disease research to identify potential biomarkers and to provide biologic insights.1 Both methodologies generate large datasets that require rigorous biostatistical and bioinformatic analysis, and both share key challenges, including variability in data quality and the potential for residual confounding which occurs in observational studies when unmeasured or inadequately controlled variables distort observed associations.