The genomes of modern humans contain faint traces of unidentified “ghost” lineages — ancient hominins that our ancestors likely interbred with long ago. Now, US scientists have pinpointed when two of these genetic signals entered the ancestral human genome, with the findings published in Science.
It has long been known that modern human DNA includes genetic contributions from other, now‑extinct human lineages, notably Neanderthals and their relatives the Denisovans — a hominin group thought to be closely related to Neanderthals and believed to have lived in East Asia, possibly until around 30,000 years ago.
There are also hints of other, more enigmatic contributors — so‑called “ghost lineages” whose identities are harder to pin down, hence the name.
According to the new research, one of these unknown ancestors appears to be a “super‑archaic” hominin lineage that split from other human lines about 1.8 million years ago and later interbred with Denisovans. The study also uncovered a second, younger lineage that interbred directly with modern humans in Africa before the last major migration of Homo sapiens out of Africa over 50,000 years ago.
“In terms of relatedness, they were very different,” says study co‑author Yulin Zhang. “The younger ghost lineage diverged from modern humans around the time of the Neanderthal split, tracing back to a common ancestor roughly 0.8 million years ago; the super‑archaic lineage is older, having separated around 1.8 million years ago. So we’re talking about populations separated from us by hundreds of thousands, or even more than a million years of independent evolution.”
The extremely ancient DNA from the super‑archaic group appears to have reached modern Homo sapiens genomes via our interbreeding with Denisovans.
This fits with an earlier study published this year that found proteins extracted from Homo erectus teeth in China carry distinctive mutations also seen in Denisovans and modern humans. Homo erectus is thought to have left Africa around 1.8 million years ago, making it a plausible candidate for this ultra‑ancient “ghost” lineage — though without direct DNA evidence, this remains speculative.
As for the younger ghost lineage, its age suggests its DNA entered the modern human genome before Homo sapiens dispersed out of Africa, meaning its genetic signature should be present in people worldwide — both in African and non‑African populations.
While earlier studies had detected signs of “ghost” ancestry from ancient African hominins in modern humans, most pointed to interbreeding occurring after the last major Homo sapiens migration out of Africa. That would mean the legacy would be limited to individuals with African ancestry.
The current study indicates that our direct ancestors did interbreed with this younger ghost lineage at some point, though the precise nature of that relationship — when, where, and how long it lasted — remains unclear.
“For the younger ghost lineage, the mixing happened before the out‑of‑Africa migration, so the signal is visible in all populations,” Zhang explains. “But we couldn’t pin down the exact timing or determine whether it was a single pulse of gene flow or a more continuous exchange.” She adds that the genetic contribution is estimated at roughly 0.5 to 1 % per individual — similar to the Neanderthal contribution, which ranges from 1 to 4 % in non‑African populations.
These results rely on a computational method called TRACE (TRacking Archaic Contributions via ARG Estimation), developed by a US research team led by population geneticist Priya Moorjani.
In addition to identifying these two episodes of interbreeding with archaic human groups, TRACE successfully recovered the well‑known cases of mixing with Neanderthals and Denisovans, confirming that these unusual genetic segments are not random noise but real evidence of encounters with unknown hominins.
“These segments carry the fingerprints of a genuinely ancient lineage: they are long, contiguous stretches that show markedly greater divergence from modern humans than the rest of the genome, plus elevated heterozygosity,” Zhang says. “These are the same hallmarks seen in confirmed Neanderthal and Denisovan segments, and they’re incompatible with random noise.”
The super‑archaic hominin lineage linked to the Denisovans was detected by analysing only the genomes of populations from Oceania, where people are known to carry high levels of Denisovan DNA — sometimes up to 4 % of their genome.
This is no coincidence. Because this DNA is thought to have entered the modern gene pool through ancient interbreeding with Denisovans, it is most clearly visible in populations that today retain the largest proportion of Denisovan ancestry.
The team believes that future research may help clarify the identity of these ancient ancestors, especially as global genomic databases grow more diverse and include broader representation of human populations.
Finding more Denisovan genomes would help, as would obtaining more protein sequences from Homo erectus fossils, the researchers note. Ideally, fossil evidence showing unambiguous proof of interbreeding would settle the matter, though such a discovery is considered unlikely.
About the Author
