Research Paper Volume 17, Issue 1 pp 170—202

Age-invariant genes: multi-tissue identification and characterization of murine reference genes

class="figure-viewer-img"

Figure 3. Classical and novel RG performance in aging samples. (A) Aging RG status of classical reference gene by tissue. Some tissue-RG pairs are age-invariant (blue), and are therefore good RG candidates for aging studies, while others did not satisfy our filtering criteria (red). (B, C) Coefficient of variance (B) or age correlation (C) violin plots of valid classical RGs (green), novel pan-tissue age-invariant RGs, and invalid classical RGs (red). (D) Scatterplot comparing %CV and age correlation for each tissue-RG pair. (E) Scatterplot comparing RT-qPCR Gene RefFinder score and mRNA-seq %CV in heart and liver. RefFinder and %CV scores were calculated from in-house and public validation datasets respectively. RefFinder is a summary score of BestKeeper, NormFinder, GeNorm and comparative delta-Ct values (analysis of these scores can be found in Supplementary Figure 5). Circled points indicate novel age-invariant RGs (Two pan-tissue: Tomm22 and Srp14; and one heart and liver age-invariant gene: Atp6v1f) while uncircled points specify classical RGs from Figure 3A. (F) Violin plots for RefFinder scores comparing invalid classical RGs and novel pan-tissue age-invariant RGs. Unless specified, p-values are obtained from a Welch Two Sample t-test (**p < 0.01, ***P < 0.0001, exact p-values in text).