Onpode
Cover art for A decades-old theory that evolution protects essential genes just got overturned

A decades-old theory that evolution protects essential genes just got overturned

November 22, 2026 · 7 min

Eleanor Crane & Ben Okonkwo

A study by Dr. Sofia Radrizzani and Professor Laurence D. Hurst at the University of Bath found that essential genes in E. coli and baker's yeast do not mutate less than non-essential genes, overturning a 50-year-old hypothesis proposed by Edward Cox in 1972. Natural selection, not reduced mutation rates, explains why vital genes appear protected.

Researchers at the University of Bath's Centre for Evolution have published a study in PLOS Biology (dated November 12, 2026, per SciTechDaily; November 19–21, 2026 per ScienceDaily) challenging a hypothesis first proposed in 1972 by evolutionary biologist Edward Cox.

0:006:54
Get the next episode on Science →

Follow it free — new episodes land in your feed.

Or make your own — any topic, in minutes

More Onpode episodes on Science →

About this episode

In 1972, biologist Edward Cox proposed that organisms might evolve lower mutation rates in their most essential genes — the ones where damage is typically fatal. It was an elegant idea, and for decades almost nobody stress-tested it. Now a research team at the University of Bath, working with what they describe as the largest dataset of spontaneous mutations assembled to date, has found no evidence for it. Essential genes in E. coli and baker's yeast mutate just as often as non-essential ones. The episode works through what the study actually did and didn't show. The key methodological challenge: a mutation that kills the cell never reaches the sequencer, so you have to catch mutants before selection removes them — and the authors acknowledge their data are only probably free of that bias. The finding is real but bounded: two microbes, not all of life, and no confidence intervals are available to confirm that "no detected difference" means a measured zero. The more interesting question the episode turns over is why the old evidence looked so convincing for so long. The answer is a little unsettling — natural selection was erasing its own tracks, leaving earlier researchers to count survivors and mistake that filter for a cause. The episode also flags which outlets picked up the story and why that matters: a result confirming how selection works was being read, in some corners, as evidence against it. It isn't. What fell is one side hypothesis. What survived is the standard picture, slightly cleaner.

Frequently asked

Do essential genes have lower mutation rates than non-essential genes?

A University of Bath study by Dr. Sofia Radrizzani and Professor Laurence D. Hurst found no evidence that essential genes in E. coli and baker's yeast mutate less than non-essential genes. The study, published in PLOS Biology, used what the authors described as the largest collection of spontaneous mutations assembled to date.

Why did scientists previously think essential genes mutated less?

Earlier researchers observed fewer mutations in essential genes, but this was most likely because natural selection rapidly removes harmful mutations in those genes before they can be counted — though this selection-bias explanation has not yet been directly demonstrated. The mutants died before reaching the sequencer, making vital genes appear better protected than they actually are.

What was Edward Cox's 1972 hypothesis about mutation rates?

Edward Cox proposed in 1972 that organisms may evolve suppressed mutation rates specifically in essential genes — those required for basic survival — because damage there is typically fatal. This idea went largely untested for decades. A recent University of Bath study found no supporting evidence for it in E. coli or baker's yeast.

Does this study disprove Darwin's theory of evolution?

The University of Bath study does not disprove Darwin's theory of evolution. It overturns only a specific side claim — that essential genes receive lower mutation rates. The broader mechanism of natural selection acting on genetic variants is explicitly affirmed by the study, making the standard evolutionary picture cleaner, not weaker.

Does the finding apply to plants and animals, or only microbes?

The University of Bath study tested only two microorganisms — E. coli and baker's yeast, Saccharomyces cerevisiae. Whether essential genes in plants and animals also lack reduced mutation rates remains untested. The authors' conclusion cannot yet be extended across all of life.

Grounded in 4 sources
Do Vital Genes Mutate Less? Huge Bacteria and Yeast Study Says No - Around Prague ↗ · aroundprague.cz
New study challenges fifty year old theory of evolution | Parallax ↗ · parallax.news
A 50-year-old theory about evolution may have been wrong all along - ScienceDaily ↗ · sciencedaily.com
Scientists Just Challenged a 50-Year-Old Theory About How Evolution Works ↗ · scitechdaily.com
Read transcript

Ben Okonkwo: Hey, you ready for this one? I've got a fifty-year-old idea that I keep wanting to be true, which is annoying for someone who's supposed to be skeptical.

Eleanor Crane: Well, that's an itch. Tell me the idea, and I'll tell you if I want it to be true too.

Ben Okonkwo: Edward Cox, 1972. He proposed that organisms may evolve suppressed mutation rates in essential genes. Those are the genes required for basic survival, where damage is typically fatal.

Eleanor Crane: Mm. So think of a builder. You reinforce the load-bearing walls and let the decorative trim take its chances.

Ben Okonkwo: Yes, exactly that. A cell copies its vital instructions more carefully. Which means mutation rates differ by gene, against the classical view of uniform, undirected mutation.

Eleanor Crane: And nobody really tested it?

Ben Okonkwo: Largely not, for decades. Too lovely to question, I think.

Eleanor Crane: Well, if nobody tested it for decades, somebody eventually did. Who, and what did they actually do?

Ben Okonkwo: Dr. Sofia Radrizzani and Professor Laurence D. Hurst, at the Centre for Evolution, University of Bath. They published in PLOS Biology. And they assembled what they describe as the largest collection of spontaneous mutations to date. In E. coli and in baker's yeast, Saccharomyces cerevisiae.

Eleanor Crane: And then, what did they see?

Ben Okonkwo: Nothing. Which is the finding. Essential genes did not mutate less often than non-essential genes. No protection. The load-bearing walls got copied just as sloppily as the trim.

Eleanor Crane: Hold on. "Largest ever" is a big phrase. Who says so?

Ben Okonkwo: The authors. That's their own description, and I haven't seen anyone verify it. Right, and size isn't really the point. What matters is the design. You have to see mutations before selection edits them, because a mutation that kills the cell never reaches the sequencer.

Eleanor Crane: So the whole thing hangs on catching the mutants before they die.

Ben Okonkwo: Yes. And the authors say the data are only probably free of selection effects. Okay, that's the assumption I'd lean on hardest. Also, two small things. Sources disagree on the date. SciTechDaily says November 12, ScienceDaily says the 19th to 21st. And it's one bacterium and one yeast. That's not all of life.

Eleanor Crane: But that leaves me puzzled. If there was no protection, why did the old evidence look so convincing? People saw something. Genes that matter really did show fewer mutations.

Ben Okonkwo: Right, and here's the answer. Natural selection removes harmful mutations in essential genes fast, before anyone can observe them. So those genes only appear to mutate less. The mutants are gone before the count.

Eleanor Crane: Hold on. So the thing being studied was erasing its own tracks.

Ben Okonkwo: Yes. And that's my favorite part, the test isn't an experiment on cells. It's a reanalysis of which mutations show up. Same selection that hid them from earlier researchers is what the paper says explains the mistake.

Eleanor Crane: Let me make it a person. A postdoc, say, on a night shift in a sequencing core, building a model of how resistance emerges. What goes wrong for her?

Ben Okonkwo: She pulls variation from clinical isolates and treats it as the mutation rate. But if selection already filtered that variation, she's measuring survival, not mutation. Her predictions of how fast resistance emerges drift off, and she'd never see why.

Eleanor Crane: Radrizzani called the earlier claims "a triumph of hype over substance." Harsh. But is the bias explanation actually shown?

Ben Okonkwo: No. Hm, well, it's proposed as the parsimonious one, not directly demonstrated here. Explaining why the old evidence misled isn't the same as showing it did. And the data are only probably free of selection effects, so that's still a good hypothesis, not a demonstrated one.

Eleanor Crane: And the arc has a hole, doesn't it? A renewed debate in 2023, and nobody says what it was.

Ben Okonkwo: Right. The sources never explain it, so 1972 to now isn't as tidy as it sounds. And there's a detail coming that makes this sharper, about when protection actually happens, and how outlets blur one dead hypothesis into something bigger.

Eleanor Crane: So if the hole in the timeline is real, what exactly has fallen? I want the narrowest honest answer, because I suspect it's smaller than the headlines.

Ben Okonkwo: It is smaller. One hypothesis fell: that genes get a lower mutation rate because they're essential. Gene-level suppression. That's it. The protection still exists, it just happens after the mutation, when selection removes the harmful variants. Mutation proposes, selection filters.

Eleanor Crane: Proposes, filters. Mm.

Ben Okonkwo: Picture a mail sorter, say, in a dead-letter office. Every letter arrives, good or garbled. Nothing is stopped at the door. The bad ones just never get delivered. Someone counting only delivered mail would swear the garbled ones were rare.

Eleanor Crane: Which is lovely, and slightly alarming. Because here's what I noticed. Some outlets carrying this, discovery.org and icr.org, have clear ideological perspectives. And the paper explicitly affirms natural selection.

Ben Okonkwo: Hm. I hadn't looked at who was carrying it.

Eleanor Crane: Well, that's my worry. A result about how selection works gets read as selection failing. One dead hypothesis, and suddenly it's the death of evolution. Is that a fair reading of what's on the page?

Ben Okonkwo: No. Not even close. The broader theory is affirmed, selection acting on variants. Cox's idea was a side claim about mutation supply. Losing it makes the standard picture cleaner, not weaker.

Eleanor Crane: And the claim itself? Does it hold everywhere?

Ben Okonkwo: Right, that's where I'd rein in even the friendly version. E. coli and baker's yeast. Two microbes. Plants and animals are untested. So "directly contradicting" is strong. And I'd want confidence intervals, because no detected difference isn't a measured zero.

Eleanor Crane: Plants and animals are still untested. Does mutation stay blind to how important a gene is out there too? Nobody can say yet.

Ben Okonkwo: Right, that one's open. And the other one, will anyone directly test the selection-bias explanation? Nothing I've seen says someone is. Until then, it's the better story, not the shown one. But I'll say, a story about protected genes can outlast its evidence for fifty years, and I understand why.

Eleanor Crane: Mm. And that small thing, we both buy. The vital instructions were never guarded. Selection simply left the survivors to count.

A decades-old theory that evolution protects essential genes just got overturned · Onpode