A human, an octopus, and a coral share more than you'd guess

Pick any three animals that look nothing alike, a human, an octopus, a coral, and you'd assume their DNA has nothing in common either. You'd be wrong. Buried inside each of their chromosomes are recognizable fragments inherited from a single shared ancestor that lived more than 600 million years ago, long before anything resembling a modern animal existed. What's new isn't that fact itself, researchers have known about these deep genetic ties for a while. What's new is a study published in Science Advances by an international team led by scientists at the University of Vienna, which for the first time compared thousands of animal genomes at once and found that the chromosomal changes separating those three creatures didn't happen randomly. They followed a limited set of fixed, one-way routes the researchers are calling "evolutionary highways."

That's a striking claim to make about hundreds of millions of years of evolutionary history across the entire animal kingdom, and it rests on a dataset large enough to back it up.

The scale that made this possible

The team analyzed more than 5,800 chromosome-scale genomes spanning 4,454 species across 19 animal phyla, according to the published study, making it the largest comparison of its kind ever assembled. That distinction, chromosome-scale, matters enormously here. Most of the genomes sequenced to date are what researchers call "draft" genomes: they list which genes an organism has, but not how those genes are physically arranged along complete chromosomes. A draft genome is like having every word from a book shuffled into a pile; a chromosome-scale genome is the same book with every page still bound in order. You can't trace how chromosomes fused, split, or rearranged over deep time without the second kind, and until recently, chromosome-scale assemblies simply weren't available at this scale.

Darrin Schultz, who led the work as a postdoctoral researcher at the University of Vienna and is now an assistant professor at Lehigh University and Lehigh Oceans, described the payoff of finally having enough data. "For the first time, we can see thousands of genomes on a single map and trace the unique paths along which animals' DNA evolved," Schultz said, adding that viewing the full map reveals patterns invisible in any single genome comparison. The team built a new analytical framework for this purpose, which they call evolutionary genome topology, projecting that entire diversity of animal life onto one comparable map.

The mechanism behind the highways

At the center of the findings is a process the researchers named "fusion-with-mixing." When two ancestral chromosome segments fuse together in a lineage, the genes from each segment don't stay neatly separated on either side of the join. Over generations, they permanently intermingle, scrambling together in a way that can never fully unscramble itself. That irreversibility is what turns these events into reliable historical markers. Once a fusion-with-mixing event happens and the genes shuffle together, there's no path back to the pre-fusion arrangement, which is precisely why the researchers describe these routes as one-way highways rather than two-way roads.

Working from a reference set known as BCnS ancestral linkage groups, essentially chromosome building blocks traced back to the last common ancestor of bilaterians, cnidarians, and sponges, the team checked 406 mathematically possible combinations of chromosome fusion pairs. They found evidence for 397 of them, either fully fused or fused-and-mixed, somewhere across the animal species they examined. That's a striking hit rate, and it means nearly the entire theoretical space of chromosome combination has actually been explored somewhere in 600 million years of animal evolution, just not all in the same lineage at the same time.

Why "irreversible" is the important word

Most evolutionary traits can, in principle, reverse or converge from different directions; a trait lost in one lineage can sometimes look similar to the same trait independently regained in another. Chromosome fusion-with-mixing doesn't work that way, and that distinction is what gives this framework real predictive teeth. Because the mixing step permanently erases the boundary between the original chromosome pieces, researchers can look at a modern genome and read its fusion history like a one-directional timestamp, the genomic equivalent of tree rings that can only add up, never subtract. That's a different kind of evidence than comparing DNA sequences directly, which can mutate at wildly different rates across lineages and sometimes converge by coincidence. Chromosome architecture changes far more rarely, but when they do happen, they leave a mark that stays legible for hundreds of millions of years.

This kind of deep, structural evidence complements other recent findings about how far back researchers can now trace fundamental features of animal life, including work suggesting that some of biology's earliest cellular transitions happened more than once independently rather than along one single path.

What this means beyond the map itself

The practical value here extends well past satisfying curiosity about deep ancestry. The researchers explicitly framed their findings as a foundation for animal biodiversity conservation, and the logic follows naturally from what fusion-with-mixing actually measures. If chromosome architecture evolves along a limited, mappable set of routes rather than unpredictably, that map becomes a tool for understanding which lineages are closely related at a structural level even when their outward appearance diverged dramatically, and for identifying evolutionary lineages that may carry unusually distinctive or vulnerable genomic architecture worth prioritizing for conservation attention.

It also reframes how researchers might think about future evolution. If the number of viable routes through chromosome-architecture space is genuinely limited, as this study's near-complete coverage of possible fusion pairs suggests, then evolution has fewer live options available to it than a fully random model would predict, a constraint that echoes other recent discoveries about surprisingly fixed evolutionary pathways, such as the single growth-line clue that pushed back the origins of live birth in ancient cynodonts by tens of millions of years once researchers knew precisely what evidence to look for.

A map still being filled in

The team's 5,800-genome dataset is enormous by any prior standard, but it still represents a tiny fraction of the roughly 2.5 million described animal species alive today, and an even smaller fraction of the estimated total that remains undescribed. Chromosome-scale sequencing is expensive and technically demanding compared to draft sequencing, which means the map the University of Vienna team built will keep expanding as more species get the higher-resolution treatment. Each new chromosome-scale genome added to that map is less a data point than a fresh confirmation, or occasionally a genuine surprise, about which of these one-way evolutionary highways a given lineage actually traveled. Given how much of that map depends on data that simply didn't exist a decade ago, the more interesting discoveries here may still be the ones this framework hasn't found yet.