Universal transcriptomic signature of age reveals temporal scaling ofCaenorhabditis elegansaging trajectories

Abstract

To detect overlap or convergence among the diverse genetic pathways that can extend lifespan, we collected a dataset of 60 C.elegans age-dependent transcriptomes by RNA-seq technique for worm strains with vastly different lifespans. We selected four exceptionally long-lived mutants and three examples of the most successful life-extending RNAi treatments (which increased mean lifespan by 35% rather than 120% as reported). We used the dataset augmented with publicly available gene expression datasets to produce a transcriptomic signature of biological age. We introduced a transcriptomic measure of biological age and observed that its dependence on chronological age is modulated by a single parameter, the rate of aging. We hypothesized that the scaling revealed in the gene expression kinetics underlies the recently observed scaling of the survival curves in C.elegans, and the stochasticity in gene expressions leads to deceleration of mortality with age, reaching a plateau at advanced ages. Using experimental survival data, we confirm that the plateau mortality agrees closely with the estimate of Gompertz exponent at the cross-over age near the mean lifespan. The genes associated with aging in our data are enriched with the targets of transcription factors such as DAF-16, ELT-2, ELT-6, NHR-10, ZTF-9, NHR-86, and miRNAs including miR-57, -59, and -244, which is in agreement with previous studies. Overall, our meta-analysis results are consistent with a concept of aging based on critical dynamics of molecular level variables (e.g., gene expression), and support our view of aging as arising from dynamic instability of a single (critical) mode.

LINT‐Web: A Web‐Based Lipidomic Data Mining Tool Using Intra‐Omic Integrative Correlation Strategy

“…These two linear projection algorithms are widely used in large “omics” datasets to conduct dimensionality reduction and clustering in unsupervised and supervised manners. [ 31–34 ] Users can download the PCA or OPLS‐DA score plot once the dimensionality reduction analysis is finished (Figure 2C and Figure S1B, Supporting Information). In addition, LINT‐web provides hierarchical clustering analysis on the “omics” datasets to plot heatmaps (Figure 2D).…”