Entropy and Epigenetics in Aging Science with Peter Fedichev and Jan Gruber — The VitaDAO Aging Science Podcast

9 min readApr 7, 2024


In the current episode of The VitaDAO Aging Science Podcast, we explore the fascinating intersection of entropy, epigenetics, and aging with our esteemed guests, Peter Fedichev, founder of Gero and a trailblazer in longevity research, and Prof. Jan Gruber from Yale-NUS, known for his deep understanding of the physics behind aging. As we navigate through Peter Fedichev’s recent paper that sparked heated discussions on the limits of age-reversal, we’ll delve into the science of stochastic changes in methylation patterns, the controversial debate around the reversibility of aging, and the impact of entropy on human longevity. This episode will also shine a light on the vital role of VitaDAO in science funding, the challenges faced by PhD students in today’s economic climate, and the exciting potential of naked mole rats in aging research.

A brainstorming session about science funding and the impact of entropy on human aging — a talk with Peter Fedichev and Jan Gruber

Peter Fedichev is considered by many of my colleagues as an extraordinarily creative and productive aging researcher. This is one of two reasons why I invited him on the VitaDAO aging podcast. The second is that Peter’s recent paper attracted a lot of controversial debate on the longevity subreddit (1) and I was hoping we could clarify some of the issues that people raised.


In his paper Peter claims that stochastic changes in methylation patterns impose a hard limit on our ability to reverse aging (2). In this podcast we discussed this thesis and whether reprogramming refutes his ideas.

We also talked about the hardships of PhD students, the role of VitaDAO in science funding, linear and non-linear features that change with age, the difference between hallmarks of aging and proximal causes of aging, the mean-field effect, stable and unstable animals, naked mole rats and more. Although ultimately we may have failed to really break down and simplify his work for the general audience, the journey was still a worthwhile one and I hope you will enjoy our podcast, even though the second part is quite technical.

Below I will provide a few comments that might be useful to better understand the podcast.

Brief bio — Peter Fedichev and Jan Gruber

“Peter Fedichev, Ph.D., is an entrepreneur and scientist with over 20 years of experience in academic research and biotech business. He is a founder of Gero, a longevity startup focused on developing therapies that will extend a healthy human lifespan. Peter’s scientific background lies in the field of condensed matter physics, biophysics and bioinformatics…His dream is to defeat aging and experience life in space.”


Prof. Jan Gruber (Yale-NUS) is another physicist turned biologist with space travel ambitions. I invited him because he is very familiar with Peter’s work in the hope that he could explain the math better than I can. Not only is Jan quite familiar with the physical ideas behind Peter’s research but he is also one of the most knowledgeable aging researchers I have met.

Science funding and PhD life

Peter thinks that VitaDAO should hire more experts and delegate decisions to them. Both Jan and Peter suggested to fund risky projections at an early stage. You can listen to the podcast to for the whole discussion covering more of the nuance here.

Another conversation we had was about the life of PhD students and we all probably agreed that the funding situation is rather dire, needing improvement.

“economically the life of a PhD student is pretty hard these days… the system is broken” (Peter Fedichev)

In that context it is interesting to note that VitaDAO realized the extent of this problem and now offers small amounts of money to young and/or low-income students and researchers. These are called the VitaDAO fellowships. The idea is that a small sum can make a big difference to someone who is in need, making it a very effective program. As far as I can tell, these fellowships are partly supported by external donations allowing VitaDAO to focus most of their internal funds on funding research while still doing community building together with other donors.


Epigenetics and epigenetic clocks

Genes encode the information to produce various proteins. Genetics deals with genes while epigenetics describes the regulation of gene expression through epigenetic marks. There are two major types of epigenetic marks which can affect gene expression. Histone marks and DNA methylation. The latter is usually attached to so called CpG islands (which are stretches of Cytosine and Guanine bases) and the object of study in most clock papers. In general, CpG methylation is thought to reduce the expression of nearby genes. However, more interestingly, it appears that these CpG methylation patterns change in certain ways with age, thus allowing us to predict a person’s age based on their DNA methylation.

What is entropy and how does it relate to aging?

Simply put, entropy describes the amount of disorder in a system. The amount of entropy in the universe increases over time, which is equivalent to the second law of thermodynamics. Similarly, the amount of entropy increases in a closed system and will thereby limit its functionality. Since biological systems are complex any increase in entropy ultimately poses an issue for their functioning.

However, the earth is considered an open system so the limitations imposed by the physical concept of entropy do not apply. On the other hand, there is a closely related concept of stochastic damage which is very similar and, indeed, such damage is considered a major driver of aging. So in way the idea that entropy drives aging was never really in doubt.

I would say the major differences in opinion are related to the number and heterogeneity of these lesions that comprise molecular damage. Optimists like Aubrey de Grey would argue that there are only few types of damage that limit human lifespan and the robust lifespan extension effects seen with compounds like rapamycin may also be taken as support for the malleability of aging.

What makes Peter Fedichev’s work interesting is that it entails a rather pessimistic interpretation instead. In the case of epigenetic aging, given the large number of CpG islands that all show stochastic or entropic changes over time it would very difficult to intervene and reverse this damage. It is not physically impossible to reverse that damage at so many different genetic loci, but technically exceedingly difficult — to put it mildly.

Perhaps inspired by the great Russian writer Peter explained that each aged cell will accumulate entropic changes, but each cell will have different changes. As they say: “All happy families are alike, but every unhappy family is unhappy in its own way” (Leo Tolstoy)

While perhaps unpopular among Biohackers and Longevity Advocates, there is strong support for the idea that epigenetic changes are stochastic. Indeed, Björn Schumacher and David Meyer from CECAD just published a manuscript showing that this stochastic variation is sufficient to construct aging clocks. While I have not read it, I do respect their work and presume that it is solid (3).

Nevertheless, Peter is optimistic that we could at least radically slow aging, as he explains in the podcast. I would speculate that perhaps compounds such as rapamycin can only slow aging and bona fide rejuvenation is indeed difficult.

Does reprogramming and evolution prove that entropy is irrelevant?

“But my biggest issue is that I can’t square their claims against the fact that the cells in our body are millions of generations old. For women and men, they generally have been alive for 20+ years before their gametes are used/generated for the next generation. Now these cells maintain their DNA in a more highly regulated way, but doesn’t this throw cold water on entropy after aging has occurred?” (Poster 1)

“Now, I have said I don’t necessarily agree with the paper. And that is because we have empirical evidence, and Dr. Sinclair is on the record saying, that there is a “backup copy” of lost information somewhere in the cell.” (Poster 2)

“I don’t really have theoretical objections to this like most of the respondents here do, but purely empirical ones: aging has already been reversed in several organisms. So there is obviously something wrong with the theory,” (Poster 3)

Indeed, evolution shows that entropy does not limit the existence of life per se, since complex cellular organisms have existed almost since the dawn of time.

Moreover, researchers have found that epigenetic reprogramming during early embryogenesis could contribute to rejuvenation of the species. Reprogramming here means the erasure and resetting of epigenetic marks which is reflected in a reduction of epigenetic age. Many commenters have argued that the success of such reprogramming, both partial in the lab and natural in the womb, implies that Peter’s thesis of irreversibility must be wrong. Hence epigenetic aging must be reversible.

Without going into much detail, I do think they have a point. However, most people seem to forget another major ingredient necessary for rejuvenation of the species’ bloodline, which is selection. Just to give a few examples, men produce hundreds of millions of sperm, which are in competition with each other for general and especially mitochondrial fitness because they have to race towards an egg cell to fertilize. This competition will help to filter out sperm cells with high entropy because these will, most likely, show reduced swimming ability, reduced ability to penetrate the egg’s zona pellucida, reduced ability to produce viable offspring, etc.

In fact, we have strong empirical evidence for selection as a complementary mechanism that together with reprogramming should reduce epigenetic age (and entropy). The almost exponential increase in fetal chromosomal abnormalities seen with the mother’s age proves that reprogramming is not enough (4). In nature, spontaneous abortions, failure to fertilize and implant — which can be caused by chromosomal abnormalities among other issues — would select against offspring with low functional status. It is tempting to speculate that a fetus or embryo that, by chance, inherited a higher epigenetic age would be selected against, at some stage.

This is my own thesis and Peter discusses some other arguments in the podcast.

What are stable and unstable organisms?

As far as I understand, in his framework Peter categorizes organisms in two groups based on their dynamic stability. When looking at physiologic traits like complete blood counts, humans show a low auto-correlation of these values whereas mice show high levels of auto-correlation. This means that a perturbation in mice would persist for longer, whereas humans are able to return to equilibrium quicker; i.e. they are more stable (5, 6).

“we observed that the fluctuations of physiological indices in humans are also dominated by a collective variable characterized by a relatively long but finite auto-correlation time (in the range of a few weeks) and associated with age and all-cause mortality. The number of individuals exhibiting signs of the loss of dynamic stability (measured by exceedingly long auto-correlation times) increased exponentially with age at a rate matching the mortality doubling rate from the Gompertz mortality law” (6)

One implication is that the effects of anti-aging treatments should persist for a long time in mice, whereas the effects in humans would be transient and small to begin with, at least during midlife which he characterizes as a period of stability.

This idea of mice being dynamically unstable is conceptually similar, although distinct, to the broader notion of mice being fragile and short-lived. You could perhaps say it is another line of evidence in favor of this idea. When we say mice are short-lived we do not just mean it in the trivial sense. Of course they are! What we generally like to do is to compare the lifespan of animals to the expected lifespan based on their body mass which is also known as the longevity quotient. Small species tend to live shorter lives. Mice, however, are even shorter-lived than expected based on their size, which is also the reason why all of us are so interested in naked mole rats that are the opposite — unusually long-lived for their size.


1. “Aging clocks, entropy, and the limits of age-reversal”

2. Tarkhov, Andrei E., Kirill A. Denisov, and Peter O. Fedichev. “Aging clocks, entropy, and the limits of age-reversal.” bioRxiv (2022): 2022–02.

3. Schumacher, Björn, and David Meyer. “Accurate aging clocks based on accumulating stochastic variation.” (2023).

4. https://www.preimplantationgeneticdiagnosis.eu/pgd/risk-of-aneuploidy-and-maternal-age.aspx#:~:text=However%2C%20the%20frequency%20of%20aneuploidy,50%25%20of%20embryos%20are%20abnormal.

5. Avchaciov, Konstantin, et al. “Unsupervised learning of aging principles from longitudinal data.” Nature Communications 13.1 (2022): 6529.

6. Pyrkov, Timothy V., et al. “Longitudinal analysis of blood markers reveals progressive loss of resilience and predicts human lifespan limit.” Nature communications 12.1 (2021): 2765.




VitaDAO is the world’s first decentralized intellectual property collective, funding and commissioning research into human longevity.