DNA from scraps at Denisova Cave and nearby sites describes a yak relative that split off 400,000 years ago, survived two ice ages in the Altai, and vanished after 27,000 years ago.

The bones were not identifiable. That is the point worth starting with: the fossils that produced this discovery were fragments too small and too nondescript for anyone to say what animal they came from by looking at them. They sat in collections from Denisova Cave in the Altai Mountains of southern Siberia and nearby sites, catalogued as unidentifiable bone.

Researchers at the Centre for Palaeogenetics, a joint venture of Stockholm University and the Swedish Museum of Natural History, sequenced them. Writing in Current Biology, they report 20 ancient genomes spanning 200,000 years, and a species nobody had described: a large bovine, related to modern yaks, that followed its own evolutionary path for hundreds of thousands of years.

“This is a remarkable discovery of a large extinct species based on small non-diagnostic bones,” said Love Dalén, professor of evolutionary genomics at the Centre and the study’s senior author.

An echo of the Denisovans

Dalén draws a comparison that will land with anyone who follows palaeogenetics. “These findings mirror the discovery of the Denisovans, a group of archaic humans, from a single fossil in 2010 and highlights the Altai region as a possible biodiversity hotspot and refugium across Ice Ages,” he said.

The parallel is exact in method. The Denisovans were identified from a finger bone at the same cave, by sequencing rather than by morphology, and are known primarily as a genome rather than as a skeleton. The same cave has now produced a second species defined the same way. Whether that reflects something special about the Altai or about Denisova Cave’s exceptional preservation conditions is a question the release raises as a possibility rather than settles — a cave that yields usable DNA from 200,000-year-old fragments will produce discoveries that a cave with poorer preservation cannot, regardless of what lived nearby.

Four hundred thousand years of separation

The genomic dating places the lineage’s divergence from modern yaks beginning around 400,000 years ago, with full separation around 250,000 years ago.

Alexandre Gilardet, a PhD student at the Centre and one of the study’s lead authors, offers a reference point. “A useful comparison for when this lineage originated is the divergence between brown bears and polar bears,” he said. Those two species are also thought to have split roughly 400,000 years ago — a span over which two animals became distinct enough to occupy different worlds while remaining capable of interbreeding.

Gilardet also notes what the climate was doing. “This age overlaps with a particularly strong warm climatic period, which could have prompted population isolation and, ultimately, speciation,” he said. A warming interval can fragment the range of a cold-adapted animal, stranding populations in high or northerly refuges where they diverge in isolation. That is a standard mechanism, and the timing is consistent with it.

A bottleneck 120,000 years ago

The 200,000-year span of the dataset is what makes the next finding possible, and it is unusual. “The long time span of our dataset is really rare in ancient DNA,” said Jonas Oppenheimer, a postdoctoral researcher at the Centre and the other lead author. Most ancient-DNA studies work with a narrow slice of time; a series of genomes across two glacial cycles allows a population’s history to be read rather than inferred.

What that series shows is a population replacement during the last interglacial, accompanied by a loss of genetic diversity. “This could reflect the genetic consequences of climate warming around 120,000 years ago on a cold-adapted species,” Oppenheimer said.

The last interglacial was warmer than the present day in much of the northern hemisphere. A cold-adapted animal passing through it would have seen its habitat contract, and a population that survives a contraction emerges with less genetic variation than it had. The genomes record exactly that: one population largely replaced by another, and diversity reduced.

There is a cautionary quality to this. The species got through that warming — it persisted for another 90,000 years at least — but it came out of it genetically poorer, and reduced diversity leaves a population less able to withstand the next shock.

No interbreeding, and then nothing

The genomes show negligible evidence of interbreeding with other bovines in the region, including steppe bison and aurochs, both of which were present. That matters for the species claim: large mammals that share a range frequently hybridise, and a lineage that stayed genetically separate despite the opportunity has a stronger case for being a distinct species.

The youngest specimen is about 27,000 years old. After that the record stops. The release states plainly that the timing and causes of the extinction are unknown, which is the honest position — 27,000 years ago is the last time the animal is known to have been alive, not the date it died out.

That period does place it in familiar company. The last glacial maximum and the millennia around it saw the disappearance of much of the Eurasian megafauna, amid both severe climate swings and expanding human populations. Which of those pressures bore on this animal, and in what proportion, is not something 20 genomes can answer.

Not yet a name

The species has no formal name, and the researchers are clear about why: genetic evidence alone is not sufficient to name a new species under zoological convention, which requires a type specimen and a morphological description. The team plans to compare their material with previously described yak-like remains from the region, some of which may turn out to belong to this lineage.

That is a real possibility worth holding. Fossil bovines from Eurasia have been named and renamed repeatedly on fragmentary material, and it would not be surprising if this genome eventually attaches to a species someone described decades ago from a partial skull.

Funding and interests

Funding came from Vetenskapsrådet, the European Research Council, Knut och Alice Wallenbergs Stiftelse and the Russian Science Foundation.

The conflict-of-interest statement deserves direct attention. Beth Shapiro is employed by Colossal Biosciences, the company pursuing de-extinction of the mammoth and other species; Love Dalén, the senior author, sits on that company’s Scientific Advisory Board. Both hold stock options. The release mentions Shapiro only in this statement and does not describe her role in the study.

Nothing about that arrangement undermines the genomics, which stands or falls on the sequence data. But a commercial de-extinction venture has an evident interest in the discovery of extinct cold-adapted megafauna with sequenced genomes and living close relatives, and readers are entitled to weigh that. The disclosure is properly made; it is simply worth reading.

---

Notes

1. Gilardet, A., Oppenheimer, J., Dalén, L., et al. “Ancient genomes reveal an extinct bovine species from mid-latitude Asia.” Current Biology, 5 October 2026. DOI: 10.1016/j.cub.2026.09.043.
2. Stockholm University. “Previously unknown extinct bovine species discovered in Siberia.” EurekAlert!, 5 October 2026. https://www.eurekalert.org/news-releases/1145741.