Site icon Scientific Inquirer

New Method Maps DNA Inherited From a ‘Ghost’ Human Lineage in Every Living Person

3D illustration of intertwined DNA strands unzipping and replicating inside a cell

A detailed 3D visualization of DNA strands during the replication process inside a cell.

UC Berkeley researchers used ancestral recombination graphs to find traces of an unknown archaic group that interbred with ancestors of modern humans in Africa, and of a 1.8-million-year-old lineage that mixed with Denisovans.

The human family tree is more tangled than it once seemed. Over the past 15 years, genetic studies have shown that the ancestors of living people interbred with Neanderthals and Denisovans, and that a few percent of many people’s DNA comes from these archaic relatives. Scientists have also suspected that other, unidentified groups left their mark, but finding those traces without a fossil genome for comparison has been difficult.

Researchers at the University of California, Berkeley, now report a method that can map such hidden contributions in modern genomes. In a study published in Science, they use the technique, called TRACE, to identify DNA from a “ghost” lineage that interbred with the ancestors of modern humans in Africa, and from an even older, “super-archaic” group that mixed with Denisovans in Eurasia.

“We were actually able to find and map genomic locations in modern humans that are from this ghost lineage,” said Yulin Zhang, the study’s first author.

Reading history from genealogies

Most earlier methods for detecting archaic ancestry work by comparing modern genomes with ancient ones sequenced from fossils, such as Neanderthal and Denisovan bones. That works when a reference genome exists. For groups known only from indirect evidence, or not known at all, there is nothing to compare against.

TRACE takes a different approach. It relies on ancestral recombination graphs, a way of reconstructing the family history of every stretch of DNA in a sample of genomes. Because chromosomes are reshuffled each generation through a process called recombination, different segments of a person’s genome have different ancestral histories. An ancestral recombination graph traces how each segment’s lineage branches and merges back through time.

Segments inherited from a deeply divergent archaic group stand out in these reconstructions because their lineages split from those of other humans unusually long ago. By looking for that signature, TRACE can flag DNA from archaic sources even when no ancient genome is available.

The study was carried out by Zhang, Arjun Biddanda, Sarah Johnson, Colm O’Dushlaine of the genomics company 54gene and Priya Moorjani, a geneticist at UC Berkeley.

A ghost in Africa

Applying the method to modern genomes, the team found evidence of a ghost lineage that diverged from the ancestors of modern humans about 800,000 years ago and later interbred with them in Africa, before about 50,000 years ago. Because the mixing happened before modern humans spread out of Africa, traces of this lineage are found in people everywhere.

The researchers estimate that this ghost group accounts for roughly 0.5 to 1 percent of every person’s genome. Across all sources, about 2 percent of the modern human genome is archaic in origin, according to the study.

Who the ghost population was remains unknown. Africa has a rich but patchy fossil record of human relatives, and the continent’s hot, humid climates are generally poor for preserving ancient DNA, so no genome from such a group has been recovered. The new results give scientists a genetic portrait of this lineage even without a body to attach it to.

Fossils show that Africa was home to a variety of human relatives during the period in question, including Homo naledi in South Africa, which lived around 300,000 years ago, and other forms whose relationships to modern humans remain debated. Linking a genetic ghost lineage to any particular fossil group would require additional evidence that the new study does not provide.

“We often think of human evolution as a branching tree,” Moorjani said. The findings add to a picture in which branches repeatedly rejoined, more like a braided stream than a neatly forking tree.

An even older lineage

The method also uncovered a deeper layer of mixing. The team found evidence that a super-archaic lineage, diverging roughly 1.8 million years ago, interbred with the ancestors of Denisovans in Eurasia. That date is close to the time when early members of the genus Homo, such as Homo erectus, first spread out of Africa.

According to the study, Denisovans carried about 3 to 5 percent super-archaic DNA. Because some modern populations, particularly in Oceania, carry up to 4 percent Denisovan ancestry, fragments of this ancient lineage persist in living people as well, passed along secondhand.

Biddanda, now a postdoctoral researcher at Johns Hopkins University, called the super-archaic result “particularly exciting,” because it reveals a genetic contribution from a lineage that otherwise left little trace in the record.

Useful inheritances

The archaic segments were not scattered at random. The researchers found that they were enriched in genes related to immunity and metabolism, suggesting that some of the inherited DNA was useful and was favored by natural selection.

That fits a pattern seen with Neanderthal and Denisovan DNA, which has contributed genes affecting immune responses, skin and adaptation to high altitude. Archaic groups had lived in their environments for hundreds of thousands of years, and interbreeding may have offered modern humans a shortcut to adaptations already tested by local conditions.

“Adaptation to new pathogens and food sources has been one of the strongest selective pressures in human evolution,” Moorjani said.

Open questions

Methods based on ancestral recombination graphs depend on statistical reconstructions of genetic history, and the dates and proportions they produce come with uncertainty. Other explanations, such as long-lasting structure among ancient African populations without a distinct ghost group, have been proposed for some similar signals in the past, and the debate is likely to continue.

The study was supported by the U.S. National Science Foundation. By making hidden ancestry visible without a fossil reference, TRACE could help researchers look for other unknown contributors to human genomes, particularly in Africa, where most of human genetic diversity resides and much of the human story unfolded.


Notes

  1. Zhang, Y., et al. “Recovering signatures of archaic hominin introgression using ancestral recombination graphs.” Science, 2026. DOI: 10.1126/science.aef8874.
  2. University of California, Berkeley. “New technique pinpoints human DNA inherited from ‘ghost’ ancestors.” EurekAlert!, 30 July 2026. https://www.eurekalert.org/news-releases/1137804.
Exit mobile version