All Your DNA Would Reach Jupiter a Hundred Times Over

Lex Fridman opens the second conversation with a simple prompt: what is beautiful about the human epigenome? Manolis Kellis, professor at MIT and head of the Computational Biology Group, does not hesitate. "Don't get me started," he says, and then he starts. Every cell contains about two meters of DNA, compacted into a nucleus whose radius is a thousandth of a millimeter. That is six orders of magnitude: as if a string as tall as the Burj Khalifa, about a kilometer, were crushed into a ball a millimeter across.

Stretch the DNA in all of a body's cells and the length is planetary. An adult has about 30 trillion cells; two meters in each of them would reach all the way to Jupiter, a hundred times over. Almost none of that length is even the interesting part of the body. The nuclei are a rounding error in volume, and Kellis calls the compaction itself the boring structural half of the story. The functional half is what the packing is for.

Humans have roughly 20,000 genes. Every cell needs only a few thousand of them, and a different few thousand depending on whether it is skin, heart muscle, or neuron. The DNA sequence is the same; identity is remembered by the epigenome. Being Greek, Kellis notes that *epi* means "on top of": the genome is the DNA, and the epigenome is everything laid on top of it. There are chemical marks on the DNA itself, marks on the histones that wrap it, and the three-dimensional folding of chromatin.

The most familiar DNA mark sits on cytosine. Add a methyl group and you have what is sometimes called a fifth base. Kellis's metaphor is an orchestral score: the notes are fixed, but annotations tell you to play a passage softly, loudly, or not at all. A methylated C in a regulatory region is basically an instruction to play that passage softly—a sign of repression. The 23 chromosomes are 23 volumes of an encyclopedia. A cell that "plays the piano" might open chapter seven of chromosome 23 and chapter four of chromosome two and never touch the rest.

The epigenome, in that picture, is the conductor. It does not rewrite the notes. It coordinates which instrument plays which excerpt, so that 30 trillion copies of the same library produce a working body instead of a pile of identical cells. Fridman loves the musical language. Kellis insists it is not decoration: every cell really is performing a different part of the same score.

Fridman asks the question that bothers anyone who has watched evolution lectures: selection scores the whole symphony, yet every part is doing an infinitely specific little job. How does that bookkeeping work? Kellis goes further than the organism. A nucleotide is tested at the level of a gene; genes and regulatory elements are tested at the level of a cell; cells make a tissue; tissues make an organism. The organism is not the end of the stack. It lives in an environment of other individuals, other species, whole ecosystems. Selection, he says, runs on all of those layers at once.

That stacked view is his answer to altruism. Why are people nice to members of their own group? Species that were simply nasty to one another did not survive as species. Plants like carbon dioxide and humans like oxygen; they trade gases. An organism that killed everything around it would starve. The pressure is not only "can I live" but "can we live together." Fridman wonders whether the same logic might run from the quantum scale to the galactic. Kellis does not need that leap. Even the word *environment* already has layers: sunlight and pollution outside the body, and thousands of coevolved genes inside it.

Move an African individual's gene into a European cellular context, he asks, and will it still work the same way? Probably not. It was ground against thousands of other genes. Neanderthal alleles make the same point. In a Neanderthal genome a given variant could be fine; dropped into a modern human genome it can be disease-causing. Genes are not spare parts. They carry a history, and they only make sense in the company they evolved with.

Neanderthals were only one of about five branches Kellis is confident about after the out-of-Africa events. Denisovans are another; almost everything known about them once came from a fragment of one pinky bone from one cave in Siberia, found less than ten years before this conversation. Homo floresiensis, the so-called hobbits of Indonesia, is yet another. Those branches vanished. Reading Greek myths as a child, the age of heroes and Titans sounded ridiculous; as an adult he wonders whether such stories were fed by bones that looked almost human, the way dragon tales may have been fed by dinosaur bones.

What killed the other branches—war, or genetics? Kellis does not think modern tribalism is far from the traits that got humans here. You back my team, my party, my village, and you are kin; you are from the next village, and I hate you; we both move to the city, and suddenly we are regional brothers. Two neighboring countries fight to the death, then meet in a third country and become compatriots. The inconsistency is logically absurd and genetically efficient. Kin selection rewards liking those who resemble you. Transport that instinct to Ice Age Europe, with Neanderthals already on the ground, and the move is ugly: they are not like us, clear them out. Territory plus resemblance is a bad combination. Something that almost looks like you is what you try to eliminate first.

The genomes tell the other half. There was plenty of gene flow: war and romance at once, Romeo and Juliet across a species line, including mixing inside Africa among growing tribes. It is a shame, Kellis says, that Earth no longer has a neighboring intelligence that is like us and yet different. Dogs and wolves found a way to live beside humans, but he doubts the wolves chose it. Humans were overpowering; keep the mild pup, kill eight of nine, and in a few generations you have a tame dog. Neanderthals were probably useful. They were also scary, and too close.

Are any living species smarter than humans? Kellis mentions dolphins: look at their brains, the way they play, the way they learn. The hardware looks sufficient. They lack opposable thumbs, and that probably mattered. The sharper point is definitional. If dolphins are super-smart, have figured out the meaning of life, and spend the day playing in water—which might *be* the meaning of life—we would never know, because all they appear to be doing is playing in water, the way sharks do. Intelligence is hard to score unless a species goes out of its way to demonstrate it.

That problem explodes if the life is alien. Fridman had recently talked with Sara Seager about looking for life on other worlds. You might detect that something is alive. How do you ask it whether it is intelligent? Kellis notes that the same question already applies to every species on Earth. His confidence is narrower: if we met a human-like intelligence, and if neither side was busy annihilating the other, he has no doubt we would decode their language in a few weeks. Sophistication of behavior and language, on our side, is the bet.

He also knocks the human body plan. Go diving and the sea is full of appendages, round things, five-way asymmetries, outrageous colors. Tetrapods are a tiny, boring minority down there. It was those few four-limbed fish that crawled out and spawned every "boring" body we see on land: four-limbed humans, four-limbed birds, four-limbed lizards. "We are that unimaginative branch," is the spirit of the rant. Dolphins, he adds, went back into the water. Maybe they were the smart ones.

If Kellis had to put all his money on where life started, he would take the seafloor, not the surface. The outside-in story is tides: moon and sun stirring nutrients. The inside-out story is hydrothermal vents, black smokers and white smokers pouring energy from the planet's core. Life needs energy. It does not need the sun. Panspermia—a meteorite delivering DNA—he finds a bit lazy, because it only postpones the question of how life began on the rock. Earth already had the ingredients. Why not here?

He wants three things. Metabolism, from the Greek for transformation: take energy and matter from the environment and make them part of yourself. Compartmentalization: without a self there is no evolution; you have to know where you end. Lipid bilayers form almost for free in water. Replication need not be self-copying. A makes more B, B makes more C, C makes more A; any closed reinforcement will do. Once that loop closes, he will not say all hell breaks loose—"all paradise breaks loose." You can modify, improve, and hand the rest to natural selection.

Early life was not DNA. DNA is a later storage medium. RNA could store information and fold into three-dimensional enzymes; that is the RNA world. Its masterpiece is the ribosome, a factory that translates RNA into protein. Proteins, with 20 amino acids rather than four bases, can take more shapes, including reverse transcriptase. Textbooks treat transcription, DNA to RNA, as the default and reverse transcription as the oddball. RNA came first, so reverse transcription was the earlier invention. RNA invents proteins; RNA and proteins together invent DNA; then RNA, in Kellis's telling, gets tired and delegates storage to DNA and most actuation to protein.

Europa, Jupiter's smallest Galilean moon, has an iron core, a rocky exterior, a thick ice shell, and likely a huge liquid ocean underneath. Tidal heating from Jupiter stirs the seafloor. The ice is shielding, like an atmosphere; there is even a thin layer of oxygen. "All the ingredients are there," he says. His guess is that an independent origin is already teeming. He is not afraid of it. Environments that different will not know how to metabolize each other; Earth bacteria would die there, and Europan chemistry would die here. Put it on a sandwich? Taste receptors and immune systems would not even recognize the molecules. Independence Day, with a Mac talking to alien hardware, is hilarious to him. Macs and PCs already struggle.

Shipped to Europa as a field scientist, what would he look for? Life, once it appears, changes the world around it in ways physical law alone would not predict. It compartmentalizes, pushes things away, keeps things inside; those traces are observable. Organizing a Meaning of Life symposium, he told an astrophysicist friend they needed biology as well as physics. She said biology is just a small part of physics. In a sense, he agrees. His reply was still no: life challenges physics, supersedes it, fights it. It does not obey entropy; it gathers energy. It does not obey diffusion; it builds a border. It does not obey gravity; it grows the other way. A chemical signature that is obviously picking that fight would be life.

Then the next layer. If life is a fight against physics, the brain is a fight against life. Evolution shaped bodies to reproduce. A brain can refuse. We can decide not to have children, not to eat, to suicide, to cut off the outside world—all the things that make a human, we can decide not to do. Kellis wants that seen as a continuum of transformations, not a jump. When Alexander the Great was asked whether he owed his life to teachers or parents, he said parents gave him *zoe*, life itself, and teachers gave him *eu zoe*, the good life. Zoe is being alive. Eu zoe is living as a person.

Getting from one to the other took leaps: compartmentalization, the eukaryotic nucleus, multicellularity, the split between germline and the rest of the body, then a coordinating brain. The rest of you, he says bluntly, exists to get that germline into the next generation. Everything else is packaging. "Sorry to be so blunt." A worm is a mouth, an anus, and some organs in between; humans are a fancier version with limbs and sense organs as tools for getting food. He jokes that worms are Windows 95 and we are a later release with a few extra features. Specialization and cooperation keep going deeper. Eu zoe is what you get when those layers include altruism and a mind that can say no.

Gene-expression responses run on minutes to half an hour: a stimulus, a search along a long string of DNA, a product. Daily life does not. Someone throws you a ball and you catch it in milliseconds, with no gene-expression change at all; there is no time. Life is stacked hardware supporting a control system that lives on a different clock. Trees look dumb until you speed a film from spring to October: they reach, grab, grow into the light. All judgments of intelligence, Kellis says, are timescale judgments. Human childhood is stretched by neoteny—babies useless for months, then years, then still in college—keeping the brain plastic. Intergenerational gaps have already moved from something like 16 years to 40. If lifespans became 400 or 800 years, a three-generation trip to another planet would be a joke.

Asked offline whether he wants to live forever, his answer is absolute. There are kinds of forever. Live today forever, Groundhog Day style? Yes: the day's to-do list would take a lifetime. Live forever while the world keeps changing? Even better, never boring. Fridman offers the opposite thought experiment: wake each morning ten years later. That would make you, among Earthly life, one of the slowest. Kellis likes that trees may have it right; olive trees have stood since Minoan civilization. A friend's genetics paper, he notes, found little outside gene flow from Minoans and Mycenaeans into modern Greeks. Those civilizations transformed thought—rules in nature, standards of beauty, the bet that the world is elegant. To him that is a species-level leap on the order of evolving a brain, language, or eyesight.

Fridman had just been at a Neuralink event where Elon Musk talked about brains as noisy memory recorders, and joked that this very conversation might already be a replay. Kellis laughs: if anyone's memory gets replayed it will be Musk's; who is he, some unknown. If he got to pick a memory to replay, he would pick talking with Fridman. The timescale talk is not an escape from mortality. It is the setup for what comes next: once the tests end, how do you inhabit the time you actually have.

The Meaning of Life symposium was an inside joke because 42 is the answer in *The Hitchhiker's Guide to the Galaxy*. Part of the motive was a midlife crisis. Kellis tells an academic rat-race joke in the spirit of Christopher Hitchens: as an undergrad you work like a rat to reach grad school; in the Ph.D. you work like a rat for a postdoc; as a postdoc, for assistant professor; as faculty, for full professor; and by the time you are full professor, you are a rat. At 42 he had the house, three kids, a beautiful wife, tenure, excellent students, piles of grants. He looked back and thought: that was an awesome race. But I am not a rat. I got out of the labyrinth.

The tests had been well-defined, with someone else's solution at the end. Real life is not a test. The next trap would have been making sure the kids were first in class, then the right college, then waking up at 60 still in the maze. Friends at the symposium advised keeping unachievable goals—progress without arrival—which he hears as more obstacles for the inner rat. His own move is a joke he shares with his wife: pretend they live in an all-inclusive resort. They have hired the people on the esplanade to bike and rollerblade for their viewing pleasure, and they feed them. What is on today? He has signed up for professor activities, a lineup of extremely smart people and a grant-writing meeting. She has consulting. In the evening they ask how the activities went.

If he were a billionaire, he would pay a great university for an office and pay brilliant people to work with him even if they did not really want to. That is the life he already has, working himself to the bone because it is so fulfilling. The rat race is destination-shaped: how do I get through the maze. The resort is path-shaped: today's schedule could not be better. He quotes C.P. Cavafy's *Ithaca*: hope the road is long, full of adventure; when you arrive, Ithaca has nothing for you; Ithaca gave you the journey, and without it you would not have set out. Destination thinking spends everything on the end. Path thinking spends it on the day.

Fridman asks about the dark stretches. Kellis is glad, because society, he thinks, does a terrible job of preparing people for pain. Every Hollywood movie needs a happy ending. You can count the bad-ending films you have seen on ten fingers and not use them all. The hero is beaten and beaten, and you still know the good guys win. We need more movies where the bad guys win, more movies where MacGyver fails to defuse the bomb and it simply explodes, more honesty that life sometimes sucks and that is all right. Growing up in Greece he heard songs that were not merely sad but miserable: a mother at dawn mourning a son killed in civil war by fellow citizens, wishing he had died with honor beside a general against a foreign army.

The children's stories are worse. One is four pages: a poor child in the street envies the rich child in the window, the rich child envies the poor child's freedom to run, and both die—one of cold, one of illness. The next: a woman whose brother is dying in a war against the Turks prays to the Virgin, who offers to take someone else. She cannot choose any of the men with families waiting. She lets her brother die, because that death has honor. These are not hymns to misery. They expand emotional range. Kellis cries reading them; his face goes red. Embracing the suffering is a choice. His kids ask for the moral, and sometimes he says there is none: life does not always make sense. Crying through worse lives also shrinks your own complaints. His mother was changed as a teenager volunteering in a hospital, watching people cling to life.

He is careful with comparative misery. Viktor Frankl in the camps, Syrian children after a bombing, his own toddler whose skin peeled from spilled hot chocolate and who stayed a happy kid—the point is not that you should want a bigger car than the neighbors. Children accept the moment. People in war zones still struggle for love, meaning, and friendship, not only calories. Another story his mother told: a man asks Christ to take his heavy bag; Christ offers other bags; the man keeps his own. Other people's loads would be worse. Zoom out and the bag is simply the path. Embrace good and bad every day and what remains is joy; without both you are not a complete person. Pixar's *Inside Out* is the civic version. Joy tries to fix everything, fails, and the girl goes numb because all she can do is fake a smile. Society keeps asking, are you happy? His mother calls with that question. He wants to say: stop. I am not happy. Ask if I am fulfilled. Ask if I am suffering beautifully, struggling for something great.

Meaningful work is not a string of joyful moments. It is struggle, success and failure, people being nasty and people being kind. If you know who you are, what others say about you only teaches you about them. He told his son that after a classmate's insult. He admits his own flaw, procrastination, and tricks it: delay the mindless chores and spend the stolen time on what matters. Life does not start after the next milestone. It started a long time ago. Happily ever after is the end of the story; nothing happens after that. The struggle is the interesting part, and the journey is the meaning—not the ending kids are trained to expect.

Does he fear getting older, the first step on the way to the last? At 18 to 20 he felt invincible. Any math, logic, or programming puzzle, he saw the answer. He floated into lecture with a newspaper, looked up now and then, and pointed out the professor's errors—"a complete brat." Ten years later the fine-grained speed was gone; the brain worked differently. Twenty years later, differently again. "Oh gosh, I love the journey." Youth is combinatorial explosion on one problem. Age is borrowing from other fields, stepping outside the box, noticing there is a ceiling in the maze and asking why you are still banging on it.

He watches his own arrogance in his kids, and he watches students do things he can no longer do, better than he probably ever did. What they cannot replace is perspective: he knows which problems are solvable because he has been there, and which he will not bother with—not because he could not, but because he is no longer interested. He is looking for the ecological niche he is adapted to, the same word he used for cells specializing. At the scale of a life, a 50-year-old hiring students and postdocs to hold the big picture is doing what a worm did on evolutionary time when it grew arms: specialize, compartmentalize, amplify what you are best at. He prefers "embracing the struggle" to "acceptance," which can sound like you cannot change anything. You shape the journey.

If you are a clump of molecules, where in spacetime would you rather sit? Not a rock on Mars, not a black hole, maybe a supernova if you did not have to survive it. Earth in this solar system is a strong draw. Being an aware animal is better; being human, with introspection, better still; being young, healthy, and surrounded by people you eat and argue with is luck. The last bump on that local maximum is kindness and gratitude. He repeats a prayer from their earlier episode: thank you for the good given me; give me strength to give others the same love. Kindness is selfishly rational—happier neighbors raise your pool—and it is also selected because giving feels unreasonably good. Fridman says the claim that Boston is the best place in the universe is itself a gravitational field. Kellis agrees: it is not only cognition. It molds the people around you, and then you really do live in a better place.

Fridman says Kellis still has an undergraduate's eyes; strangers would not guess the size of the lab. As a new professor Kellis tried adult clothes: khaki pants, patterned shirts, a proper belt. A few months later he could not stand it, and went back to rollerblades, T-shirt, and baseball cap. Colleagues' CVs were terrifying; he assumed a mistake had been made. The correction was a bias. He saw others' work as flawless and his own as a third of what he had meant, full of holes. They saw only the finished third, which was already good. He maps it onto a bad definition of machine intelligence: whatever machines can do stops counting as intelligent, so the territory of "real" intelligence shrinks. As soon as a kid like him could solve a problem, he tagged it easy. Colleagues' problems still looked impossible. Both sides were misreading the same landscape.

Marvin Minsky's habit of never being satisfied is, Kellis thinks, a stage he has left. Maturity is accepting that the work helped someone else step up, and therefore moved human knowledge. The trap of "everything I have done is crap" is not wisdom. What he tells students is that their best work is ahead. What they have is great for now and will be obsolete in 30 years; the next thing will be better. Perfect cannot be the enemy of good. In a meeting a student said the paper was not perfect; Kellis said it was already better than anything in the field, and if they kept polishing it would never ship. Get it out, then improve it. In his experience they never do the improvement, and that is fine: you placed a stepping stone and others will pass you. His graduate adviser was harsher. After the first paper, do not write the second. Give the field away. Do not sit on a summit firing papers two through five because it is comfortable. Shocking at the time, then liberating. You do not have to outrun everyone. You put a stone down; together people cross a wider river than one giant leap.

Alex Stark, already full of first-author papers, was interviewing with Rick Young and Eric Lander when a friend sent him to first-year faculty Kellis. Kellis rollerbladed into the Stata amphitheater, did not take the skates off, talked gene regulation for half an hour, and left. Stark later emailed: he wanted to postdoc with Kellis. He did not care about the costume. He cared about the ideas and the childish excitement. Between 11 and 19 Kellis changed country or school every year—Greece, France, New York—which in retrospect let him reinvent himself because no peer group had him pinned. At the time, leaving the south of France for New York cement and snow, he was simply furious. Only later did he see the gift. MIT peers, some of them, had treated the student as a future colleague before he felt like one: respect for what you say, not for your rank.

He reads two poems written on assignment. The first, in a third-language English class at 16, after his first real snow, starts with children laughing at snow and ends with the same snow having lost its light without a you; "birds who love are birds who cry." The same fact can be read as joy or as grief. Joni Mitchell's *Both Sides Now* is the later echo. Asked about love, he barely pauses: it defines his life; he is a human first and a professor second. Passion's etymology is suffering—the Passion of Christ, Greek *pathos*, sympathy as suffering-with. He does not mean misery. He means emotional investment. Hold back, shield the heart, and love cannot enter. Simon and Garfunkel's *I Am a Rock* is the warning. Love is the fragility you hand someone, no safety net. The second poem, *The Tide Waters*, written for a yearbook goodbye, is a lake of calm that becomes a river: some find their way, some sink, the hard part is still to come, and he will always need his friends' hands. Fridman closes with Alexander the Great: nothing is impossible to the one who will try. The rest of the episode has been arguing that the trying *is* the life.