BioWire Bytes 018 - The Information Theory of Aging

Byte-sized biotech

As we get older, we start to look and feel older; the face will begin to develop fine lines and wrinkles, recovery from that night out partying doesn’t just take longer, but impacts the subsequent days in a noticeable way, the memory dulls, new sensations of pain in our muscles and joints, and the inevitable development of back pain. These are all things that happen with aging.

But why do they happen, and for that matter, what is aging? There are two ways to think of aging: chronological age, how many birthdays you’ve had, and phenotypic age, a measure of how your body’s systems compare to the averages across the lifespan. You can feel young and be old and vice versa, be young but feel old. It’s phenotypic age that captures how old your body acts.

So yes, aging looks like wrinkles, new aches and pains, and losing memory. But beneath the surface, is it just wear and tear, or is something deeper happening at the cellular level?

Researchers are beginning to think aging isn’t simply mechanical decline. Instead, it might be a communication breakdown in our cells’ software. A leading framework, the Information Theory of Aging (ITOA), suggests aging stems from the gradual loss of epigenetic information, the “software” that tells our genes how to behave, rather than irreversible DNA damage. In other words, our cells don’t necessarily break down, they forget their instructions, much like a corrupted hard drive. As Harvard geneticist David Sinclair explains: “Why do we age? The epigenome is the issue… fertilization resets the epigenome, but the good news is we’ve figured out a way to safely reset that without having to clone yourself.” This optimistic outlook, that there’s a backup of youthful information in our cells, is shifting how scientists think about aging and how we might reverse it.

---

First, if you enjoy these updates, consider subscribing and becoming a part of our growing community!

---

Aging as an Information Error

In the ITOA framework, aging is essentially an information degradation problem. Our genetic code (DNA sequence) is a robust digital archive of information, like read-only data, that generally remains stable over time. By contrast, the epigenome (chemical tags on DNA and proteins, chromatin structure, gene expression patterns) is a more fluid “digital–analog” system that cells use to regulate which genes are on or off. Analog systems are powerful for fine-tuned control, but they’re inherently prone to noise and gradual distortion. Over the years, the epigenetic “signal” in our cells gets noisier: crucial genes misexpress, cellular identity blurs, and tissues begin to malfunction. This theory elegantly explains why aging manifests in predictable ways (grey hair, weaker muscles, etc.) despite each individual accumulating different random mutations, the root cause may be the same progressive epigenetic noise corrupting the cellular program.

Claude Shannon’s 1940s work on information theory provides a useful analogy. Shannon described how, when sending a message, noise in the channel can scramble the signal, but an “observer” with access to the original message can detect errors and send corrections. Borrowing this concept, Sinclair and colleagues suggest our cells have a similar built-in “observer”, essentially a backup copy of the original epigenetic information. Aging, in this view, is like a glitchy transmission: the youthful blueprint (the message) gets distorted by life’s wear and tear (the noise), leading to an aged, garbled message. But if cells could refer back to the clean backup and restore lost data, they might regain function. As Sinclair explained in an interview, “the information theory of aging is all about the preservation of information and having a backup copy.” In other words, cells might retain the original youthful “software” and, with the right triggers, can reboot themselves to a healthier state.

Genes vs. Epigenome and Digital Stability vs. Analog Vulnerability

It’s important to distinguish between genetic and epigenetic information in this model. The genome is often likened to a digital storage, information coded in sequences of A, C, G, T letters. Copying digital info can be error-corrected; indeed, cells have proofreading and repair enzymes to fix DNA replication errors. The epigenome, however, behaves more like an analog recording overlaying that genetic code. It’s subject to gradual wear from things like environmental stresses, DNA damage, inflammation, and even just the passage of time can alter epigenetic marks in small, cumulative ways. If the genome is a pristine music CD, then the epigenome is a vinyl record that can develop pops and hiss. Over decades, the “music” (gene expression) grows distorted, even if the underlying DNA track is intact.

Notably, epigenetic changes are systematic and reproducible across individuals, unlike random mutations. For instance, DNA methylation patterns change with age so consistently that scientists have created “epigenetic clocks” to measure biological age (Bocklandt et al., 2011). Different people (and even different species) show remarkably similar epigenetic drifts as they get older. This wouldn’t be the case if aging were driven purely by random DNA damage, which suggests it’s an ordered process of information loss. In fact, organisms like yeast can grow old and die with virtually no DNA mutations, indicating that something other than genetic damage is at play. Likewise, mice engineered to carry higher mutation loads, or even cloned animals derived from the cells of old individuals, often live normal lifespans. Their genomes may be littered with errors, yet their bodies don’t necessarily age faster, implying the epigenetic “software” state is the key player in aging, more so than the genetic “hardware”.

More Evidence that Epigenetic Noise is Driving Aging

A series of recent experiments is putting the ITOA to the test and the results are compelling. Here are some of the key findings that support the idea that aging is caused by corrupted epigenetic information rather than irreversible gene damage:

• Break the “Signal,” Age the Animal: Sinclair’s team developed a clever mouse model called ICE (Inducible Changes to the Epigenome) to introduce a burst of DNA double-strand breaks that scramble epigenetic markers without introducing new mutations. The “repaired” DNA was sequence-intact, but the epigenome was disturbed – and the mice began to show aging-like symptoms: graying fur, decreased vitality, organ dysfunction, and molecular signs of old age. In essence, by adding noise to the epigenetic code, they accelerated aging, even though the genetic code was unaltered. This directly supports the idea that it’s the cell’s reaction to damage – the disarray of chromatin and gene regulation – that drives aging (Yang et al., 2023).

• Mutation Load ≠ Aging Rate: Conversely, mice with extra-fast mutation rates (or people with conditions like Werner syndrome that predispose to DNA damage) don’t always show proportional acceleration in aging. And perhaps most telling, a cloned animal created from the nucleus of an old cell is reborn as a young, healthy individual (Wakayama et al., 2013). Both mice and even advanced-age livestock have been cloned and effectively had their lifespans “reset”. This is only possible because a young embryo’s environment restores the aged epigenome to a youthful state – the genome wasn’t the barrier. Identical twins (who share the same DNA) can also age at different speeds, presumably due to divergent epigenetic changes over their lifetimes.

• Epigenetic Clocks and Predictable Decay: Researchers have identified 12+ hallmarks of aging (like telomere shortening, mitochondrial dysfunction, etc.), but many of these may be downstream effects of an upstream information loss. Epigenetic changes sit near the top of that cascade. The existence of accurate epigenetic clocks, algorithms that read DNA methylation patterns to predict age, highlights that the aging process follows an informational “program.” Slower epigenetic drift is even correlated with longer maximum lifespans in mammals. In other words, species that preserve epigenetic information better (think naked mole rats or bowhead whales) tend to live much longer, hinting that maintaining a clean signal delays aging.

• Rewinding the Clock (in Cells): Perhaps most astonishing, aged cells and tissues can be rejuvenated by resetting their epigenetic marks, without altering the DNA sequence. Old human cells can be induced into pluripotent stem cells, which effectively erases age-associated marks back to an embryonic state (an extreme reset). But even short of a full “reboot,” exposing cells to youthful factors can reverse their gene expression age. In one experiment, old mouse and human cells treated with a partial “reprogramming” protocol began to behave like younger cells, and even in living mice, aspects of aging have been reversed without fixing any underlying mutations. This suggests much of what goes wrong with age is not permanent damage, but malleable, if we can restore the original epigenetic information.

There is a story emerging here. One that paints a picture consistent with the information theory of aging, i.e., aging is not a one-way accumulation of rust, but a loss of information in coding genes. But one that is showing evidence it may be reversible, like a software slowdown that might be debugged or restored.

---

These newsletters take significant effort to put together and are totally for the reader's benefit. If you find these explorations valuable, there are multiple ways to show your support:

• Engage: Like or comment on posts to join the conversation.

• Subscribe: Never miss an update by subscribing to the Substack.

• Share: Help spread the word by sharing posts with friends directly or on social media.

---

References:

Bocklandt, S., Lin, W., Sehl, M.E., Sánchez, F.J., Sinsheimer, J.S., Horvath, S. and Vilain, E., 2011. Epigenetic predictor of age. PloS one, 6(6), p.e14821.

Guarente, L., Sinclair, D.A. and Kroemer, G., 2024. Human trials exploring anti-aging medicines. Cell metabolism, 36(2), pp.354-376.

Lu, Y., Brommer, B., Tian, X., Krishnan, A., Meer, M., Wang, C., Vera, D.L., Zeng, Q., Yu, D., Bonkowski, M.S. and Yang, J.H., 2020. Reprogramming to recover youthful epigenetic information and restore vision. Nature, 588(7836), pp.124-129.

Lu, Y.R., Tian, X. and Sinclair, D.A., 2023. The information theory of aging. Nature aging, 3(12), pp.1486-1499.

Scott, A.J., Ellison, M. and Sinclair, D.A., 2021. The economic value of targeting aging. Nature Aging, 1(7), pp.616-623.

Teschendorff, A.E. and Horvath, S., 2025. Epigenetic ageing clocks: statistical methods and emerging computational challenges. Nature Reviews Genetics, 26(5), pp.350-368.

Wakayama, S., Kohda, T., Obokata, H., Tokoro, M., Li, C., Terashita, Y., Mizutani, E., Nguyen, V.T., Kishigami, S., Ishino, F. and Wakayama, T., 2013. Successful serial recloning in the mouse over multiple generations. Cell Stem Cell, 12(3), pp.293-297.

Yang, J.H., Hayano, M., Griffin, P.T., Amorim, J.A., Bonkowski, M.S., Apostolides, J.K., Salfati, E.L., Blanchette, M., Munding, E.M., Bhakta, M. and Chew, Y.C., 2023. Loss of epigenetic information as a cause of mammalian aging. Cell, 186(2), pp.305-326.