Research Paper

Age prediction from human blood plasma using proteomic and small RNA data: a comparative analysis

class="figure-viewer-img"

Figure 2. Age-associated proteins in blood plasma. (A) Age-dependent changes for 612 proteins as measured by HRM-MS (untargeted proteomics) in 19 young and 25 old individuals. The volcano plot shows the log2 fold change in protein abundance on the x-axis and the Benjamini-Hochberg (BH) FDR-corrected Mann-Whitney test on the y-axis. Red and blue colors highlight significantly up- and down-regulated proteins, respectively (FDR < 0.2). (B) Age-dependent changes for 31 proteins measured by MRM-MS (targeted proteomics) in 103 individuals. A linear model was fitted for each protein, with age as the dependent variable and the log of protein abundance as the independent variable. The volcano plot shows the estimated coefficients on the x-axis and the BH FDR on the y-axis. Red and blue colors highlight significantly up- and down-regulated proteins, respectively (FDR < 0.2). (C) Scatter plot of the correlation with age of MRM-MS measurements (x-axis) and HRM-MS (y-axis). Blue line and shadow: linear regression and 95% confidence interval, respectively. (D) KEGG pathway enrichment analysis for significantly up-regulated proteins from the MRM-MS and HRM-MS experiments. For both experiments, the set of all measured 612 proteins was used as a background to compute the significance of age-association. Colors indicate the number of age-associated proteins that are attributed to these pathways. (E) Examples of scatter plots for two proteins detected as age-associated in the two MS experiments.