Research Paper Volume 15, Issue 2 pp 472—491

Reduced sphingolipid biosynthesis modulates proteostasis networks to enhance longevity

class="figure-viewer-img"

Figure 1. Summary of transcriptomics data. (A) Experimental Design- samples were taken every hour as indicated from vehicle (Veh) and myriocin (Myr) treated cultures. RNA was isolated and mRNA was quantified using RNA-seq. (B) Differentially expressed genes (DEGs)-RNA-seq data were filtered to include well-annotated genes with sufficient signal intensity prior to statistical analysis. Two-way ANOVA (Time: 6 time points; Drug: two conditions- Veh and Myr) was applied. The ANOVA test produces 3 p-values for each gene, one for the main effect of time (T- black circle), one for the main effect of drug (D- red circle), and one for the ‘Interaction’ between Time and Drug (I- green circle; analyzed separately- see Figure 2). DEGs (p ≤ 0.01 on at least one of the three p-values) are shown. Because each DEG could be significant by 1-3 ANOVA p-values, data are displayed in a Euler diagram to indicate overlap. (C) For DEGs significant by Time (T), Drug (D), or both (T&D), post hoc analysis categorized each DEG into one of 11 different temporal patterns (00- no temporal effect; 01 and 05- changed at the first or last time point; 02–04- plateaus at intermediate time points; 06–09- spikes at intermediate time points; 10- linear change with time; for each temporal pattern other than ’00’, a gene could be assigned to one of two horizontal reflections- ‘a’ or ‘b’), and within each temporal pattern, one of three drug effects (increased by Myr- ↑ orange, decreased by Myr- ↓ blue, not changed by Myr- ↔ black) was defined. Numbers of genes assigned to each pattern are included on the right side of each pattern diagram (* significantly more DEGs than expected by chance; p ≤ 0.01, binomial test). (D) Results for each asterisked pattern from C are shown. On the left, heatmaps of standardized average signal at each time point for vehicle and myriocin groups for 5 representative DEGs are shown. On the right, the graphed standardized averages (± SD) for all genes assigned to each pattern are shown (scale bar at bottom). Pattern names are given by the Euler diagram region, then the temporal pattern # and ‘a’ or ‘b’ designation, followed by drug effect (in parentheses) and the number of genes assigned. All data for the 5169 RNA seq transcripts that passed filtering metrics are shown in Supplementary Table 1 along with the two-way ANOVA analysis results which can be used to sort for genes in each sector of the Euler diagram.