Project Design and Overview

The scope and design of the project are outlined in Figure 1. We sought to examine changes in transcripts, proteins, and metabolites over the course of cardiac remodeling from hypertrophy (HYP) to heart failure (HF). Transcriptome and metabolome determinations were made with six biological replicates per experimental group, while quantitative proteome profiling was performed by comparing sham, HYP, and HF animals in three independent multiplexed experiments. The statistical analyses differed in accordance with prevailing conventions. Microarray analysis was performed according to the methods of Irizarry et al.¹⁵ A linear model empirical Bayesian (EB)-modified Student’s paired two-sample t test¹⁶ was applied for rigorous determination of the statistical significance of differential protein levels. Finally, as elements of the metabolome were quantified individually and on multiple platforms, differential metabolite levels were determined by a Welch’s two-sample t test. The effect of multiple-hypothesis testing on the differential regulation false-discovery rate (FDR) was documented using the q-value method of Storey.¹⁷ Detailed descriptions of data processing and statistical methodology are provided in panel 1 in the Supporting Information.

Study design.