Scientists found lost human relatives hiding in your DNA, no fossils required
A new method reads the genomes of living people and finds the fingerprints of extinct human lineages known only from the DNA they left behind. It is a real advance, and a careful one: it reveals the shape and timing of these 'ghost' ancestors far more than their names. Here is what the study actually shows.
On 30 July 2026, a team led by UC Berkeley published a method that hunts for extinct human ancestors without any fossils, by reading the genealogies hidden inside the DNA of living people. It mapped two "ghost" lineages, long suspected but never pinned down: one that split from our line roughly 800,000 years ago and left about 0.5 to 1% of its DNA in every human alive today, and a much older "super-archaic" lineage that mixed into the Denisovans and reached us only faintly, second-hand. The signal is peer-reviewed and striking. The honest limit is just as important: the method reveals these ancestors' shape and timing, but not who they actually were.
Somewhere in your genome, right now, are fragments of DNA from human relatives that science has never identified. No fossil has ever been matched to them. Just their genetic signature, passed down through hundreds of thousands of years and sitting in the cells of nearly everyone alive. A study published on 30 July 2026 in the journal Science claims to have found two of them, and it did so without digging up a single bone.
That is a genuinely new kind of discovery, and it comes with a genuinely important caveat. This is one of those science stories where the excitement and the limits are inseparable, so here is the careful version: what the study found, how it is even possible, and where the confident science stops and the informed guessing begins.
What did the study actually find?
The UC Berkeley team, led by Priya Moorjani with co-first authors Yulin Zhang and Arjun Biddanda, reported evidence for two extinct human lineages that interbred with our ancestors and left traces in modern DNA. Scientists had long suspected such "ghost" populations existed, known only by the mark they left in living genomes; what is new here is the ability to map where their DNA sits, to show that one of them is present in every living human, and to estimate when the mixing happened, all without a fossil, since no bone has ever been genetically matched to either lineage.
The two are very different, and keeping them straight matters (the numbers are easy to garble):
- A "ghost" lineage that split from our ancestral line an estimated 800,000 years ago and contributed roughly 0.5 to 1% of the DNA of every living human, in African and non-African populations alike.
- A far older "super-archaic" lineage, whose divergence is estimated at around 1.8 million years ago, that mixed not into us directly but into the Denisovans (an extinct group we already knew about). It reached modern humans only indirectly, and faintly.
How can you find an ancestor with no fossil?
This is the clever part, and it is worth understanding rather than taking on faith.
Your genome is not a single unbroken story; it is a mosaic stitched together from all of your ancestors, and different stretches of it have different ages and histories. The Berkeley team built a tool called TRACE (for Tracking Archaic Contributions via ARG Estimation) that reconstructs those hidden family trees, technically "ancestral recombination graphs," across the genomes of present-day people only. No ancient DNA, no fossils. By working out the genealogy of each segment, the method can flag pieces that are far older than they should be, the tell-tale sign of DNA that entered our line from a long-separated population.
That is the leap. For the last two decades, we learned about our extinct relatives mainly by sequencing their actual bones, the ancient-DNA revolution that revealed Neanderthal and Denisovan DNA in living people and earned Svante Pääbo the 2022 Nobel Prize in Physiology or Medicine. But that approach can only reach populations whose fossils survive and still contain readable DNA. Reading the genealogies inside living genomes, in principle, can reach lineages that left no usable bones at all. That is why this method can see ghosts.
Two ghosts, two very different stories
The most common mistake this story invites is mixing up the percentages, so be precise.
The younger ghost lineage is the one that is genuinely in everyone: about 0.5 to 1% of your genome. Because it shows up in African and non-African populations alike, the interbreeding most likely happened in Africa, before the ancestors of today's non-Africans made their final exit from the continent. That alone reshapes the picture, most attention to ancient interbreeding has focused on events outside Africa, and this points to a mixing event within it.
The super-archaic lineage is where the "3 to 5%" figure comes from, and it is crucial not to misread it: that DNA makes up 3 to 5% of the Denisovan genome, not yours. Only a small fraction of it trickled down into modern humans by way of Denisovan ancestry. So no, you are not "3 to 5% super-archaic." You carry a faint echo of a population that mixed with a population that mixed with us.
The honest caveat: we found the shape, not the name
Here is the line the breathless version of this story tends to erase. TRACE can describe these ancestors' statistical shape and timing, roughly when they branched off, roughly how much they contributed, but it cannot tell you who they were. The team is explicit about this limit. As the Berkeley announcement puts it, it is "unclear who these ghost and super-archaic ancestors were."
The tempting labels, that the ~800,000-year lineage might be a Middle Pleistocene Homo group in Africa, or that the ~1.8-million-year one might be Homo erectus, are time-overlap guesses, not identifications. They mean only that the estimated split dates line up with when those known species existed, not that anyone has matched a ghost to a species. And the dates themselves are inferred divergence times from a model, not fossil-dated events. So this is not the discovery of a new human species, there is no bone to name, and nobody has read this population's genome directly. It is the detection of a signal that something once mixed into us, described in the language of statistics rather than anatomy.
That is not a knock on the work. It is exactly what an inference-from-living-DNA method can and cannot do, and the authors present it that way. The credibility of the result rests on being clear about that boundary.
Why it matters
Even bounded honestly, this is a meaningful step. For a long time, the human family tree was popularly drawn as a ladder marching toward us. Paleoanthropology traded that for a branching bush decades ago, and the ancient-DNA era added the crucial twist: those branches interbred, making it less a tree than a braided river of populations that split, wandered, met again and mixed. This study extends that picture into territory fossils cannot reach, suggesting the braid has strands we had no idea were there, and hinting that interbreeding within Africa, not just outside it, shaped who we are.
If the method holds up and is applied more widely, it could turn "ghost" populations from vague statistical rumours into a systematic map of the relatives we absorbed and lost. It will not put faces to them on its own; that still needs fossils and, ideally, their DNA. But it changes what we can even look for. The useful way to hold this result is with both hands: a real new window into deep human history, and a firm reminder that seeing a shadow is not the same as meeting who cast it. For more from the lab, see the Science section and our explainer on what CRISPR gene editing can really do.
The two ghost lineages at a glance
| Younger "ghost" lineage | "Super-archaic" lineage | |
|---|---|---|
| Estimated split from our line | ~800,000 years ago | ~1.8 million years ago |
| Where its DNA is | In all living humans (~0.5–1% of the genome) | ~3–5% of the Denisovan genome; only a trace in us |
| How it reached us | Interbreeding in Africa, before the final out-of-Africa migration | Indirectly, via the Denisovans |
| Who it was | Unknown (time overlaps Middle Pleistocene Homo) | Unknown (time overlaps Homo erectus) |
For the first time, we can see whirlpools on the Sun's surface. They may explain why its atmosphere is so strangely hot
The world's largest solar telescope has taken the sharpest images ever of the Sun's surface and caught something long predicted but never seen there: tiny plasma whirlpools, only about 20 km across, the same wave-curl you get in ocean waves and clouds. They may be part of the answer to a decades-old mystery. Here is what was found, and what it does and does not prove.
JWST found a 'black hole star' in the early universe, and it might solve a real puzzle
The James Webb telescope spotted an object 660 million years after the Big Bang that looks like a giant star but is really a black hole wrapped in so much gas it disguises itself. If the interpretation holds, it helps explain one of astronomy's genuine mysteries: how monster black holes appeared so impossibly early. Here is the careful version.
A blood test can now help diagnose Alzheimer's. Here's what that really means
Diagnosing Alzheimer's used to require an expensive brain scan or a spinal tap. In 2026 a simple blood test crossed from the research lab into the clinic, sharply improving how accurately doctors can identify the disease. It is a genuine turning point, and also not a crystal ball. Here is what these tests do, and what they do not.