Aging manifests itself through a decline in organismal homeostasis and a multitude of 26 cellular and physiological functions 1 . Efforts to identify a common basis for vertebrate aging 27 face many challenges; for example, while there have been documented changes in the 28 expression of many hundreds of mRNAs, the results across tissues and species have been 29 inconsistent 2 . We therefore analyzed age-resolved transcriptomic data from 17 mouse organs 30 and 51 human organs using unsupervised machine learning 3-5 to identify the architectural and 31 regulatory characteristics most informative on the differential expression of genes with age. We 32 report a hitherto unknown phenomenon, a systemic age-dependent length-driven 33 transcriptome imbalance that for older organisms disrupts the homeostatic balance between 34 short and long transcript molecules for mice, rats, killifishes, and humans. We also demonstrate 35 that in a mouse model of healthy aging, length-driven transcriptome imbalance correlates with 36 changes in expression of splicing factor proline and glutamine rich (Sfpq), which regulates 37 transcriptional elongation according to gene length 6 . Furthermore, we demonstrate that 38 length-driven transcriptome imbalance can be triggered by environmental hazards and 39 pathogens. Our findings reinforce the picture of aging as a systemic homeostasis breakdown 40 and suggest a promising explanation for why diverse insults affect multiple age-dependent 41 phenotypes in a similar manner. 42
43Main text: 44
45The transcriptome responds rapidly, selectively, strongly, and reproducibly to a wide variety of 46 molecular and physiological insults experienced by an organism 7 . While the transcripts of 47