The ancestor of all living horses, donkeys and zebras lived about four million years ago, suggests a new study, pushing back the confirmed age of the horse’s progenitor by two million years.

The discovery comes from the genetic analysis of a 700,000-year old horse fossil trapped in the Canadian permafrost. That’s hundreds of thousands of years older than any genome ever sequenced before.

Among other insights, the sequence supports the often-debated view that the Przewalski’s horse, native to the Mongolian steppes, is the last living population of truly wild horses in the world.

PHOTOS: Could These 10 Animals Be Resurrected?

And while the new study offers an intriguing look into the history of horses and how they have changed over millenia, the research also opens up the possibility of getting a much longer view into the evolution of all sorts of species, including people.

“This really shows that you can go much further back in time and do genomics than people previously thought,” said Eske Willerslev, an evolutionary geneticist at the University of Copenhagen. “Suddenly, that means that we can potentially go back and do the genome for precursors to Neanderthals. Maybe there’s potential for getting the genome of Homo erectus. From a scientific standpoint, this is really great.”

In 2003, Willerslev and colleagues retrieved a horse fossil from a site in Canada’s Yukon Territory, which contains some of the oldest permafrost on Earth. Based on volcanic ash preserved in the soil, the researchers estimated that the fossil was at least 700,000 years old.

When they scanned the bone for biomolecules, they were surprised to find both collagen and proteins, giving them hope that the bone might also still contain DNA, even though the oldest surviving DNA ever recovered from a fossil to date was only about 130,000 years old.

The team started with classic techniques to amplify DNA and build a genetic library, Willerslev said, but they ran into too much contamination. Most of the DNA they extracted belonged to microbes.

Przewalski horses in Khomyntal, Western Mongolia, were imported in 2004 from Le Villaret, France.Claudia Feh, Association pour le cheval de Przewalski: TAKH, Le Villaret, F 48125 Meyrueis Tak

Instead, they turned to newer technologies to put together the full genome of the ancient creature. For comparison, they also sequenced the genomes of five domestic horse breeds, a donkey and a Przewalski’s horse, the rare wild horse that lives in central Asia.

Now that they had such an old specimen to contrast with modern versions, the researchers were able to calibrate the mutation rate and then work back to an age of origin for the ancestral horse. The Equus genus, they report today in the journal Science, emerged between four and 4.5 million years ago. Eventually, that lineage split, leading to modern Arabian horses, Icelandic horses, donkeys and related species.

NEWS: Humans Tamed Horses All Over the World

There has been plenty of debate about when exactly Equus began, and while some paleontologists have suspected a history spanning 4 million years, the oldest known true horse fossil is only 2 million years old. The new findings strengthen the idea of an older history. “This study corroborates with molecular evidence what the fossils say,” said Bruce MacFadden, a paleontologist at the University of Florida, Gainesville. “It’s always interesting and good when two methods arrive at the same conclusion.”

With the ancient horse genome now available for reference, the new study offers insights into how natural selection shaped horses over time. It also offers opportunities to figure out which traits people focused on when they domesticated horses about 6,000 years ago, said James MacLeod, an equine genome expert at the University of Kentucky, Lexington’s Gluck Equine Research Center.

“Before, it was pure speculation, like ‘Oh, I think people wanted horses that were bigger and stronger to be war horses that could pull equipment and knights in heavy armor so I bet they selected for this,’” MacLeod said. “Now you could theoretically have data to back that up.”

What’s more, the data can now finally answer a long-standing question about whether the Przewalski’s horse is actually a wild strain or if it bred with domestic horses at some point in the past. According to the new work, the team reports that the Przewalski’s horse is truly wild.

And even though Przewalski’s horses went through a genetic bottleneck when the population dropped to only about a dozen individuals in the 1950s, Willerslev said, the good news is that the strain managed to retain a healthy amount of genetic diversity.

“There’s been debate about whether these horses should be elevated to the pedestal of an ancient horse breed and a still-intact wild population and protect it as such,” MacLeod said. “These data presented in this study say yes, on a genomic level, they appear to have no evidence of cross-breeding. That’s very exciting, and it will hopefully spur commitments from people who care about horses and governments that reign in the area to protect them.”