Transcriptome of Left Ventricle and Sinoatrial Node in Young and Old C57 Mice

Figures

• 

  Figure 1:. Clustering and enrichment analyses of the four groups of samples. A. The hierarchical clustering analysis of the heart tissues. B. The PCA of the heart tissues. The LVs (LY and LO) were separated from the SANs (SY and SO) by the first principle component (PC1); the old tissues (LO and SO) were separated from the young tissues (LY and LO) by the second and third principle components (PC2 and PC3). C. Heatmap generated with the normalized FPKM values. The genes were clustered into three clusters. D-E. Bubble charts of GO biological process analysis (D) and KEGG pathway analysis (E) with genes in each cluster. The larger the bubble, the more identified genes in the term. The redder, the less adjusted p-value, indicating more significant. F. Comparison of the canonical pathways (SAN/LV) enriched within the IPA software. The orange color represents the positive activation z-score and indicates the activated pathway; the blue color represents the negative activation z-score and indicates the inhibited pathway. The deeper the color, the larger the absolute value of the activation z- score. LY: young LV, LO: old LV, SY: young SAN, SO: old SAN.

• 

  Figure 2:. Visualization of KEGG pathways whose genes were expressed more in LVs and in the cluster1 of Figure 1C. The KEGG pathways of cardiac muscle contraction (A) and oxidative phosphorylation (B) were remodeled by PathView. The color denotes the scaled average FPKM value in each group: from the left to the right, young LV, old LV, young SAN, and old SAN. The redder, the higher scaled value, indicating expressed more in the corresponding group; the greener, the lower scaled value, indicating expressed less in the corresponding group.

• 

  Figure 3:. Visualization of KEGG pathways whose genes were expressed more in SANs and in the cluster2 of Figure 1C and proteasome. The KEGG pathways of ribosome (A) and protein processing in endoplasmic reticulum (B) were remodeled by Path View. In Figure 3B, genes in the blue irregular box are involved in the protein degradation, processing proteins to be degraded in the proteasome (green circle). Then the KEGG pathway of proteasome (C) was also remodeled by Path View.

• 

  Figure 4:. Visualization of the mammals-specific longevity regulating pathway whose genes were clustered in the cluster3 of Figure 1C. A. The mammals-specific longevity regulating pathway was remodeled by Path View. B. The relative gene expressions were summarized. The blue arrow denotes the lower expression, and the red arrow denotes the higher expression. C-D. Heatmap of adenylyl cyclase genes (C) and sirtuin genes (D) in the four groups. E-F. Scatter plot of adenylyl cyclase genes (E) and sirtuin genes (F) with the q-values between two groups: *, q-value < 0.05; **, q-value < 0.01; ***, q-value <0.001. The bar shows the mean and 95% CI.

• 

  Figure 5:. Longevity effect analysis of RNA-seq data in this project. A-B. Heatmap of possible prolongevity (A) and antilongevity (B) genes by mapping our RNA-seq data to the GenAge database. C-F. Volcano plots of possible prolongevity (C and E) and antilongevity (D and F) genes in LV (C and D) and SAN (E and F). The cutoffs of the regulation are: (1) -log10(q-value) = 1.3, i.e., q-value = 0.05, and (2) absolute value of log2(Old/Young) = 1. The red dot (Up) denotes the upregulated gene in old tissues; blue (Down) denotes downregulation; grey (None) denotes non-significant.

• 

  Figure 6:. Upstream transcription regulator analysis. A-B. Prediction of upstream transcription regulators by the comparison of top 500 genes in Cluster1 and bottom 500 genes in Cluster2 in Figure 1C, using the ChIP-seq (A) and Motif (B) model of LISA database. C-D. Comparison of the upstream regulators predicted in the IPA software by the comparison of SAN/LV (C) and Old/Young (D). E-F. Scatter plots of transcription regulators genes. Transcription regulators consistently predicted using LISA (A and B) and IPA (C) are displayed in Figure E. Transcription regulators predicted by the comparison of SAN/LV (C) and Old/Young (D) are displayed in Figure F. The bar shows the mean and 95% CI. q-values between two groups: *, q-value < 0.05; **, q-value < 0.01; ***, q-value < 0.001.

• 

  Figure 7:. microRNA analysis. A. Heatmap of pre-miRNAs which were expressed in >= 3 samples in each group. B-C. Venn diagrams (left) and corresponding tables (right) of differentially expressed genes. D-E. Regulation of miRNAs on differentially expressed genes between old and young samples. Genes identified in the RNA-seq are distributed in the circle, and predicted upstream mature miRNAs are in the center. The pink dot denotes the upregulated gene in old tissues, the light green dot denotes the downregulated gene, the orange dot denotes the predicted increased miRNA, and the blue dot demotes the predicted decreased miRNA. The arrow denotes activation, and the line dotted line denotes inhibition. The orange and blue edges suggest that the gene expression changes were in accord with the predicted function of the miRNAs. If the downstream gene is upregulated, the edge is in orange; downregulation, in blue. The yellow edge suggests the discordant function.