▶

Figure 1. Mr. DNA’s lesson from Jurassic Park (Spielberg, 1993, Universal Pictures): the mosquito, the amber, and the frog DNA that filled the holes. Still from the film, used for commentary; it plays the clip on YouTube.

Later in the film, the fossil hunter Alan Grant crouches in the grass and turns a piece of eggshell over in his hand. It is hatched, and empty. Every dinosaur in the park was made female, so none of them should be able to breed. “Life found a way,” Grant says. The frog DNA brought frog instructions with it. One of them let an animal switch from female to male. The dinosaurs’ cells could not tell which parts of their DNA came from a frog. They read all of it and followed all of it.

Some of the cartoon is true. DNA is a code, written in four letters, and a living thing is built by reading it. Insects do get trapped in amber, wings and legs and hairs kept whole for millions of years. In 2013, scientists described a mosquito, 46 million years old, with traces of blood still in its belly. It was pressed in shale, a flat, flaky stone, not amber, and it held no DNA.

Real DNA would not have lasted. Moa were giant flightless birds in New Zealand. In their bones, a short stretch of DNA has a half-life of about 521 years: after that long, half of it has broken apart. The oldest DNA recovered so far is about two million years old, from ancient sediment in northern Greenland. The last dinosaurs died sixty-six million years ago.

Living DNA has no frog in it, but it does have a backup. Every letter is stored twice, once as itself and once as its partner on the facing strand: T always across from A, C always across from G. When a letter on one side is damaged, the cell reads the partner across and puts the right one back.

Mr. DNA never said who does the reading. In a living cell, three kinds of enzymes do it, far too small to see, and they work like a crew with three jobs. One checks the letters. One carries a copy of them away. One builds from the copy.

The checker has plenty to do. About 330 billion cells in a human body are replaced every day. Most are blood cells, made deep in the bones by marrow cells that copy all of their DNA every time they divide. Each copy is checked as it is made and ends up with about one wrong letter in a billion.

At your next checkup, the form on the clipboard will ask about family history. It lists diseases, cancer among them, each with a box to tick and a space for which relative had it and how old they were. Some of the cancers on it trace back to the checkers themselves, a handful of proteins built from genes like any other. In some families, one of those genes is broken from birth, wrong letters slip through, and colon cancer can come years early. Watch for three relatives with colon or uterine cancer, across two generations, one of them before fifty. That pattern points to Lynch syndrome, and a genetic test can find it. About one person in three hundred carries it, and most of them do not know. The box matters even without Lynch. Someone I know learned she had rectal cancer in her late forties. Her son had his first colonoscopy at thirty-six, about ten years younger than she was at diagnosis, and nine years before most people in the United States start. The rule is short enough to check tonight. When a parent, brother, sister or child has had colon or rectal cancer before sixty, screening starts at forty. If that relative was diagnosed before fifty, it starts ten years before their age at diagnosis.

The Inspector

The camera sweeps in low over an empty football field, grass catching early light. This is a second film, about real DNA and the checkers that catch wrong letters before they settle in. Nothing moves yet, and there is not a sound. Imagine the voice that follows as something like Morgan Freeman’s, calm and certain, the voice that says what is real.

Narrator: What does DNA look like? It is a chain of small repeating units called nucleotides. Each one carries a single base, one of four, written T, C, G or A. In this story, each nucleotide is a sprite.

The grass at the edges of the frame blurs. Its color deepens and its outlines sharpen, until the picture looks hand drawn. The film has tipped from real into imagined without announcing the switch.

The sprites appear first: thumb-sized wingless creatures with a faint shimmer across their skin, pastel one second and rainbow-bright the next. The eyes of these sprites burn with a rich, liquid amber, capturing the warmth of trapped sunlight. They look like smooth, polished resin or droplets of glowing honey. Each one holds a single glowing, magnetic bead pressed close to its chest, stamped with one of four letters: T, C, G or A.

Figure 2. The four kinds of sprite, one for each letter: T, C, G and A, each holding its bead. Drawn for this piece, using a sprite illustration from Reverso Dictionary only as a reference: no wings, new colors, and the beads added.

The field is full of six billion sprites.

Picture each sprite holding hands with the sprite to the left and to the right of it, stretching the length of the field. Then picture two strands of these sprites leaning into each other and twisting over and over. They wrap around each other the way two vines climb together, with nothing solid at the center. That twist, seen from above, is the one biology textbooks draw: the double helix. Each sprite also holds hands with one partner across the twist, the same pairs as the backup: T with A, C with G. So the two strands hold together, rung after rung, end to end. One of the two strands is brand new, copied from the other a moment ago.

Watching over the sprites is an imp, and it looks nothing like them. It is low to the ground and broad-shouldered, its face dominated by a nose too big for it. The nose twitches constantly, the one part of the imp that never stays still. It walks on its own two feet, head low, sniffing. An imp is not a nucleotide and never was. It stands for something else entirely: an enzyme, a piece of the cell’s machinery that acts on the code rather than being part of it.

The imp lifts a tiny megaphone anyway, out of habit, and calls down the field, “INSPECTION!”

This happens every time a cell copies its DNA. Most of the walk is uneventful: T holding A, C holding G, pair after pair, nothing for the imp to do but keep walking. Then, somewhere past midfield, the imp’s nose twitches harder and stops.

A T has ended up paired with a C. The copying makes slips like this now and then, far more often than the finished strand shows, because most of them get fixed. A wrong pair like this one gives off a smell, faint but unmistakable: a skunk that got too close. In a cartoon it would earn its own wavy green cloud. The imp follows its nose straight to it.

It pries the mismatched pair apart. The two beads do not resist. They were never locked, only resting uneasily against each other: no click, no jolt, nothing holding them there but bad luck. The imp waits, and within moments the real partner drifts into place, the A the T should have had all along. This time there is a click, sharp as a snap, and a jolt runs through the imp’s fingers as the right bead seats home. The imp moves on, nose twitching, for the next few billion yards.

Narrator: Real pairs are held together by hydrogen bonds, weak pulls between the bases, a little like the pull of the beads’ magnets. A T and a C fit badly and hold weakly, so a pair like that one stays loose. The smell is invented. The repair is not. Real cells run mismatch repair enzymes that follow the copying machinery, scan the new strand for loose pairs like this one, and fix them.

Narrator: These are the enzymes Lynch syndrome breaks. A carrier is born with one broken copy of one of these genes and one working copy. In any cell that loses the working copy too, the inspector stops catching slips, and they pile up. That is why carriers are usually advised to start colonoscopies in their twenties or early thirties, depending on the gene. They repeat them every year or two; for everyone else, screening now starts at forty-five. A carrier who never finds out may wait for forty-five like everyone else, a decade or more after the advised start. A positive test is news for a brother, a sister or a grown child, who may carry the same broken copy. A slip the inspector misses stays put like any other letter, waiting to be read.

The Courier

A second imp has been waiting at the sideline the whole time, and it never fixes anything. It is smaller and quicker, restless in a way the bead-carrying sprites are not, and more interested in reading than in landing anywhere for good. This one goes by Quill. Imps like Quill do not carry a bead of their own to keep. They carry an empty thread, waiting for something worth copying onto it.

The double line rests, twisted and still. Somewhere inside it is the wing-shimmer verse: a stretch of letters that holds the recipe for the shimmer on a wing. It runs about a thousand sprites from end to end, sometimes several times that, and does nothing until it is needed. Just before the verse begins, a short handful of beads, no more than a few dozen sprites long, forms a pattern unlike anything nearby. This pattern is a flag. It marks where reading starts and carries no recipe of its own. Every courier knows that pattern on sight, the way the inspector knows a smell. The moment a courier passes over the flag, it knows that everything past it is worth reading.

Narrator: A verse like this one has a real name: a gene. Most DNA codes for no protein at all, only long stretches of sprites holding hands. A gene is the stretch that gets read and built into something real: a protein, a working part of the body. That is why genes matter far more than their small share of the DNA would suggest. They turn into things a person can see, feel or measure, like the enzyme that lets a body digest food, or the protein that keeps a heart beating. The flag is real too. A short marker just before each gene, called a promoter, tells the reading machinery where to start. Whether a courier ever sees that flag is decided elsewhere, by nightcaps that put a verse to sleep.

Quill peels off from the sideline and flies straight for the flag. It leaves alone the hands each sprite holds along its own strand. Instead it slips in sideways between the two strands and pries the partners across apart, one pair at a time. It works like unzipping a jacket: the teeth open only where the pull has passed. Here the zip closes again right behind Quill, so only a few pairs are ever open at once.

With the two sides briefly open, Quill reads down one of them, letter by letter, and threads a single new strand as it goes, matching each letter to its partner. An A on the open strand pulls in a paler, imp-only bead standing in for T. A T pulls in a plain A, a C pulls in a G, and a G pulls in a C. Quill’s beads are a shade paler than the sprites’ beads, a little cooler under light. Unlike the double line it came from, the new thread gets no partner of its own. If the sprites hold beads labeled A-T-C-G, Quill writes them as U-A-G-C, with U for its pale bead. Thousand-some beads later, Quill reaches the end of the verse and snaps free, and the two strands fall back together, sealed.

Narrator: RNA uses a related letter, uracil (U), wherever DNA would use T: Quill’s paler bead. Messenger RNA is not built to last. A typical one survives long enough for the builders to read it, often hundreds or thousands of times, before the cell’s enzymes break it down.

Narrator: That is also why an mRNA vaccine cannot rewrite a person’s DNA. The vaccine is a courier’s thread delivered from outside. The body’s builders read it for a few days, the cell breaks it down, and it never goes back into the double line. So when someone at a dinner table worries that the shot changes their genes, the courier answers that worry. The better question is what the builders make from the thread.

The Builder

Quill touches down at a cluster of stockier imps, the builders, round-shouldered and built for handling. Quill hands its thread over and lifts off again. The builders are stationed where they are needed, this time at the base of a flat, colorless pair of wings. The wings belong to no sprite on this field. They belong to the much larger creature this field lives inside: the whole field is the DNA in one of its cells.

The builders read Quill’s thread three beads at a stretch, always three. Each triplet acts like a single word rather than three separate letters.

A builder reads the first triplet and reaches into a supply shelf stocked with tiny colored droplets. It pulls out the one droplet that triplet calls for, clips it onto a growing chain, and moves on to the next three beads. Another triplet, another droplet, clipped on right after the first. Each triplet calls for one droplet and no other. A wrong droplet does not fit the call and will not clip on, the same exact-match rule that decided which beads could pair.

If the inspector had missed that T and C back on the field, the builders would never know. The wrong letter would be copied into the thread like any other. Its triplet could call for a different droplet, and the builders would clip it on without a pause. Like the dinosaurs’ cells with the frog letters, they read all of it and follow all of it.

Narrator: In Lynch syndrome, this is how the cancer starts. Slips the inspector would have caught land in genes that tell a cell when to stop dividing, and the builders make those proteins wrong. The early colonoscopies are the step against it: they find growths while they are small enough to remove. Not every carrier gets cancer, and why some do young and others never do is still being studied.

Eventually a triplet arrives that calls for no droplet at all. This is the stop signal: the work here is finished. The builders let go of the chain: a long, straight row of droplets with no shine to it yet, waiting to become something more.

Narrator: The builders are ribosomes, and the droplets are amino acids. Each three-bead word is a codon, and each codon calls in one amino acid. The stop signal is real too: a stop codon tells the ribosome to let the chain go.

Then the chain curls, almost too fast to watch, like a ribbon pulled once across a scissor blade. It folds in on itself, droplet finding droplet, until it locks into one fold and stays there. The fold catches light, thin and blue on one side, warm and gold on the other, the instant it settles against the wing.

Figure 3. One short verse read through: the strand Quill reads, the thread Quill makes, and the droplet the builders clip on for each three-bead word. The verse is invented; the matching and the codon readings are real.

Narrator: The shimmer is not invented either. A butterfly’s blue comes from scales layered so finely that light bouncing between them boosts some colors and cancels others, with no dye at all. The closest real version of the wing’s shimmer swims in the ocean. Squid make proteins called reflectins, each one a gene’s worth of droplets, folded like any other. Inside the squid’s skin, reflectins pack into thin stacked plates, and the gaps between those plates set the color the squid reflects. When a nerve signal arrives, the reflectins clump tighter and water squeezes out. The plates thin, and the shimmer slides from red toward blue while a diver watches. Gene, protein, fold, stack, shimmer: that is the builder imps’ chain, plus the stacking step the story leaves out.

Figure 4. A Caribbean reef squid at Bari Reef, Bonaire, its skin shimmering with structural color. Photo by Betty Wills (2005), CC BY-SA 4.0, via Wikimedia Commons.

Narrator: Many antibiotics work by jamming the builders inside bacteria. Bacterial builders differ from human ones, so the drug stops theirs and leaves the body’s own mostly alone. A cold is a virus, and a virus has no builders of its own. It hands its thread to the body’s builders, and they read it like any other. An antibiotic finds nothing to jam. When a visit for a sore throat turns to talk of a prescription, one question decides it: is this infection bacterial?

A Closing Invitation. The dinosaurs followed the frog letters because cells read whatever they are handed, right or wrong. The inspector, the courier and the builders each come back here as one small check on what your cells get handed.

1. The next time an mRNA vaccine comes up, at a pharmacy counter or across a dinner table, picture the courier’s thread: read for a few days, broken down, never written into the double line. Which worry is left once that one is gone?
2. Before your next checkup, call one relative and ask who in the family had colon or uterine cancer, and at what age. Start with the one who always knows who married whom. Two names and two ages can move your first screening years earlier. Who else needs to hear the answer?
3. The next cold or flu will bring a scratchy throat and the pull to ask for antibiotics. Before you ask for the prescription, ask the builders’ question: is this one bacterial? Which of your last few prescriptions would pass it?

T always finds A because an inspector walks every new strand and sets wrong letters right. In some families, that inspector is broken from birth. The phone in your hand can find a broken one years early.

Where This Practice Came From

In 1895, a local seamstress told the Michigan pathologist Aldred Warthin that she would die of cancer, because so many in her family had. She did, of cancer of the uterus. Warthin traced her relatives and in 1913 published them as “Family G,” arguing that a weakness for cancer could be inherited. In the 1960s, Henry Lynch found the family again: more than 650 members, 95 of them with cancer. The pattern now carries his name. The checking itself was worked out later. Paul Modrich showed how cells repair mismatched pairs after copying, and shared the 2015 Nobel Prize in Chemistry for it. The pairing rule the sprites act out comes from Watson and Crick’s 1953 paper. It ends on a line pointing at the copying the inspector checks: “It has not escaped our notice that the specific pairing we have postulated immediately suggests a possible copying mechanism for the genetic material.”

Intellectual Honesty Note. Every mechanism the sprites act out is standard molecular biology, confirmed at each step through the Narrator’s asides. The sprites, the imps, the football field and the snap are invented staging. Real mismatch repair does not swap a single bead: it cuts out a stretch of the new strand and copies it again. The shimmer as one folded protein is also a simplification; real structural color needs many molecules stacked in layers. The field holds one copy of the genome; a real cell holds two, one from each parent. The intro’s three-relatives rule is a short form of the clinical checklist, which also asks that the relatives be closely related and counts other Lynch cancers. Wrong pairs and wrong droplets are rare, not impossible: a T and a C can hold weakly, and builders do now and then clip on a wrong droplet. The Jurassic Park science is the film’s own: only a few frog species, some reed frogs among them, have been reported to change sex.

References

Allentoft, M. E., Collins, M., Harker, D., Haile, J., Oskam, C. L., Hale, M. L., et al. (2012). The half-life of DNA in bone: Measuring decay kinetics in 158 dated fossils. Proceedings of the Royal Society B, 279(1748), 4724–4733.

Centers for Disease Control and Prevention. (n.d.). COVID-19 vaccine basics. Retrieved October 5, 2026, from https://www.cdc.gov/covid/vaccines/how-they-work.html

Centers for Disease Control and Prevention. (n.d.). Healthy habits: Antibiotic do’s and don’ts. Retrieved October 5, 2026, from https://www.cdc.gov/antibiotic-use/about/index.html

Davidson, K. W., Barry, M. J., Mangione, C. M., Cabana, M., Caughey, A. B., Davis, E. M., et al. (2021). Screening for colorectal cancer: US Preventive Services Task Force recommendation statement. JAMA, 325(19), 1965–1977.

DeMartini, D. G., Izumi, M., Weaver, A. T., Pandolfi, E., & Morse, D. E. (2015). Structures, organization, and function of reflectin proteins in dynamically tunable reflective cells. Journal of Biological Chemistry, 290(24), 15238–15249.

Greenwalt, D. E., Goreva, Y. S., Siljeström, S. M., Rose, T., & Harbach, R. E. (2013). Hemoglobin-derived porphyrins preserved in a Middle Eocene blood-engorged mosquito. Proceedings of the National Academy of Sciences, 110(46), 18496–18500.

Kjær, K. H., Winther Pedersen, M., De Sanctis, B., De Cahsan, B., Korneliussen, T. S., Michelsen, C. S., et al. (2022). A 2-million-year-old ecosystem in Greenland uncovered by environmental DNA. Nature, 612(7939), 283–291.

Kunkel, T. A. (2004). DNA replication fidelity. Journal of Biological Chemistry, 279(17), 16895–16898.

Lynch, H. T., & Krush, A. J. (1971). Cancer family “G” revisited: 1895–1970. Cancer, 27(6), 1505–1511.

Modrich, P. (2016). Mechanisms in E. coli and human mismatch repair (Nobel Lecture). Angewandte Chemie International Edition, 55(30), 8490–8501.

National Comprehensive Cancer Network. (2024). Genetic/familial high-risk assessment: Colorectal, endometrial, and gastric, Version 3.2024. Journal of the National Comprehensive Cancer Network, 22(10), 695–711.

National Human Genome Research Institute. (n.d.). Transcription. In Talking glossary of genomic and genetic terms. Retrieved October 4, 2026, from https://www.genome.gov/genetics-glossary/Transcription

National Human Genome Research Institute. (n.d.). Translation. In Talking glossary of genomic and genetic terms. Retrieved October 4, 2026, from https://www.genome.gov/genetics-glossary/Translation

Nurk, S., Koren, S., Rhie, A., Rautiainen, M., Bzikadze, A. V., Mikheenko, A., et al. (2022). The complete sequence of a human genome. Science, 376(6588), 44–53.

Renne, P. R., Deino, A. L., Hilgen, F. J., Kuiper, K. F., Mark, D. F., Mitchell, W. S., et al. (2013). Time scales of critical events around the Cretaceous-Paleogene boundary. Science, 339(6120), 684–687.

Schwanhäusser, B., Busse, D., Li, N., Dittmar, G., Schuchhardt, J., Wolf, J., et al. (2011). Global quantification of mammalian gene expression control. Nature, 473(7347), 337–342.

Sender, R., & Milo, R. (2021). The distribution of cellular turnover in the human body. Nature Medicine, 27(1), 45–48.

Shaukat, A., Kahi, C. J., Burke, C. A., Rabeneck, L., Sauer, B. G., & Rex, D. K. (2021). ACG clinical guidelines: Colorectal cancer screening 2021. American Journal of Gastroenterology, 116(3), 458–479.

Smale, S. T., & Kadonaga, J. T. (2003). The RNA polymerase II core promoter. Annual Review of Biochemistry, 72, 449–479.

Spielberg, S. (Director). (1993). Jurassic Park [Film]. Universal Pictures.

Vasen, H. F. A., Watson, P., Mecklin, J.-P., & Lynch, H. T. (1999). New clinical criteria for hereditary nonpolyposis colorectal cancer (HNPCC, Lynch syndrome) proposed by the International Collaborative Group on HNPCC. Gastroenterology, 116(6), 1453–1456.

Vukusic, P., & Sambles, J. R. (2003). Photonic structures in biology. Nature, 424(6950), 852–855.

Warthin, A. S. (1913). Heredity with reference to carcinoma: As shown by the study of the cases examined in the pathological laboratory of the University of Michigan, 1895–1913. Archives of Internal Medicine, 12(5), 546–555.

Watson, J. D., & Crick, F. H. C. (1953). Molecular structure of nucleic acids: A structure for deoxyribose nucleic acid. Nature, 171(4356), 737–738.

Wills, B. (2005). Caribbean Reef Squid colors [Photograph]. Wikimedia Commons. CC BY-SA 4.0.

Wilson, D. N. (2014). Ribosome-targeting antibiotics and mechanisms of bacterial resistance. Nature Reviews Microbiology, 12(1), 35–48.

Win, A. K., Jenkins, M. A., Dowty, J. G., Antoniou, A. C., Lee, A., Giles, G. G., et al. (2017). Prevalence and penetrance of major genes and polygenes for colorectal cancer. Cancer Epidemiology, Biomarkers & Prevention, 26(3), 404–412.