Humans Did Not Come From One Ancestor: What DNA Is Revealing About True Human Origins

There is a story of human origin most of us were taught in school. Once upon a time, in Africa, a single ancestral population of Homo sapiens emerged. Over thousands of years they spread across the continent, then left Africa around 60,000 to 70,000 years ago, eventually populating the entire planet. One family. One trunk. One origin.

That story is wrong — or at the very least, far too simple. In May 2023, a landmark study published in Nature fundamentally overturned this picture. Led by Professor Brenna Henn of the University of California, Davis, and Professor Simon Gravel of McGill University, with researchers from Baylor College of Medicine and Stellenbosch University in South Africa, it presented a new model of where we came from. Not one ancestor. Not one population in one place. But a loosely connected network of early human groups, spread across Africa, exchanging genes across vast distances for hundreds of thousands of years before they began to separate.

And that is only the beginning. Because what happened after early humans left Africa turned out to be even more tangled. They met Neanderthals. They met Denisovans. They interbred with populations we have not yet even named. The story of human ancestry is not a tree. It is a web — intricate, layered, and still being untangled by geneticists in 2026. This article covers all of it: the science, the key researchers, the debates, and what it means for understanding who we truly are.

Two of those populations, Neanderthals and Denisovans, have very different literatures. A PubMed title search on 8 October 2026, counting every spelling, returned 1,251 papers with Neanderthal or Neandertal in the title and 143 with Denisovan or Denisova — almost nine to one, for populations that both contributed DNA to living people.

The imbalance tracks the fossils rather than the importance. Neanderthals have been studied since the first fossil was recognised in 1856; Denisovans were identified from a finger bone in 2010, and only in 2025 were they matched to a nearly complete skull.

That is worth holding onto while reading the network model below. What we know about human origins is shaped by which bones happened to survive and be found, and the genetics keeps discovering populations the fossil record never recorded at all.

The Single-Origin Model and Why It Was Always Incomplete

The idea that all modern humans descend from a single founding population in Africa — the Recent African Origin model, or “Out of Africa” — is not wrong in its core claim. The evidence that Homo sapiens originated in Africa is overwhelming and remains fully intact. Every human alive today traces their ancestry to Africa. That has not changed. What has changed is our understanding of what happened within Africa during the hundreds of thousands of years before our species spread out.

For much of the twentieth century, the dominant model assumed a relatively simple process: a population of early humans in one part of Africa developed the traits of modern humans and expanded across the continent and beyond. It was elegant, intuitive, and supported by early fossils from Ethiopian sites like Omo Kibish and Herto, where some of the oldest known Homo sapiens remains — dated between 160,000 and 233,000 years ago — were recovered.

The older of those dates is itself new. Until 2022 the Omo Kibish skull was usually given as about 197,000 years old; a re-dating of the volcanic ash lying above it pushed its minimum age to about 233,000 years. And in 2017 new excavations at Jebel Irhoud in Morocco, at the opposite end of the continent, produced Homo sapiens remains dated to about 315,000 years ago. Early members of our species were spread from north Africa to the east well before 200,000 years ago, which is hard to square with a single cradle in one region.

But geneticists working with modern DNA kept noticing patterns the simple model could not account for. African populations showed levels of genetic divergence from one another that were far too complex — and far too ancient — to fit a single-founder story. There were two possible explanations: either our ancestors had interbred with unknown archaic hominins, or the early human populations themselves had been more spread out, more numerous, and more interconnected than anyone had modelled. The 2023 Nature study showed it was primarily the latter — and the distinction matters enormously.

To understand how DNA carries this ancestral information across hundreds of thousands of years, it helps to appreciate how much is packed into the genome — three billion base pairs, copied with extraordinary fidelity generation after generation, leaving traces of every population encounter and split in the patterns of variation geneticists can now read.

The Nama People: The DNA That Changed Everything

At the heart of this research is a community most people outside southern Africa have never heard of: the Nama, an Indigenous pastoralist group living in Namibia and the Northern Cape of South Africa. The Nama belong to the broader Khoe-San group — Indigenous southern Africans who speak distinctive click-consonant languages and who, genetically, carry one of the oldest and most distinct human lineages on Earth.

Research published in Molecular Biology and Evolution in 2020 by Carina Schlebusch of Uppsala University and collaborators, who sequenced 25 genomes from five Khoe-San groups, found that the Khoe-San harbour the greatest genetic diversity of any group of people in the world. Fully 25 percent of their genetic variants are unique to their populations, found nowhere else in the human gene pool. A February 2026 study in Nature Communications, led from the University of Sydney, generated whole-genome data from 150 Khoe-San individuals across 12 groups and identified over 1.3 million previously unknown single-nucleotide variants. It dated the San lineage’s divergence to about 115,000 years ago and found that the San and the Damara are separate deep lineages, each of which contributed a wave of ancestry to the Nama.

How deep is that split? The estimates disagree widely. A 2017 study in Science by Schlebusch and colleagues, using ancient genomes from KwaZulu-Natal, put the first divergence among modern humans between 350,000 and 260,000 years ago. The 2023 network study put the earliest split among living populations at 120,000 to 135,000 years ago, and the 2026 catalogue at about 115,000. The 2023 authors argue that much of this spread comes from fitting the wrong kind of model: if early populations kept exchanging genes, a single “split date” is partly an artefact of the assumptions used to calculate it.

For the 2023 study, Eileen Hoal and Marlo Möller of Stellenbosch University helped collect saliva samples from 44 Nama individuals between 2012 and 2015. Those 44 newly sequenced whole genomes became the critical new data — the key that unlocked a continent-wide picture of human origins no previous study had possessed the diversity to see.

The Weakly Structured Stem: A Network, Not a Tree

The name the researchers gave their best-fitting model — the “weakly structured stem” — does not sound like a revolution. But the concept is one of the most significant shifts in our understanding of human prehistory in decades. Before modern human population structure existed as we recognise it — before there were genetically distinct groups in East, West, and southern Africa — there were two or more loosely connected populations of early Homo living across the continent. They were genetically similar but not identical, and crucially, never fully isolated. They moved, they met, and they mated, exchanging genes across vast distances over hundreds of thousands of years.

The earliest detectable split between the ancestors of contemporary populations occurred between 120,000 and 135,000 years ago, when the lineage ancestral to the Nama and other Khoe-San groups began to diverge. But even after that, migration and interbreeding continued. The stem was “weakly structured” precisely because its roots were never cleanly separated. Aaron Ragsdale of the University of Wisconsin-Madison, who contributed the key demographic-inference methods, led the team in testing a comprehensive range of competing models — and the weakly structured stem consistently outperformed all others, including models that attributed African diversity to interbreeding with archaic mystery populations.

The numbers are striking: only 1 to 4 percent of the genetic differentiation between contemporary human populations can be attributed to drift between the ancestral stem populations. The early groups were far more similar to each other than an isolated-founding-population model would predict. Their diversity was shaped not by separation but by connection. Think of it not as a family tree with one trunk, but as a river delta seen from above — multiple channels flowing in parallel, occasionally merging, occasionally separating, but always in contact before eventually diverging into the distinct streams we recognise as modern human populations.

This matters for the same reason that epigenetics has changed our view of inheritance — both reveal that biological inheritance is more interconnected and dynamic than the clean textbook models suggested.

The Neanderthal Question: How One Conclusion Was Overturned in 13 Years

In 1997, geneticist Matthias Krings, working in Svante Pääbo’s laboratory in Munich, extracted mitochondrial DNA from the original Neanderthal type specimen — the bones found in Germany’s Neander Valley in 1856. It was one of the most technically demanding feats in the history of molecular biology. What Krings found was clear: Neanderthal mitochondrial DNA fell completely outside the range of variation seen in any modern human. According to the Smithsonian Institution’s Human Origins Program, Neanderthal and modern human mtDNA differed by about 27.2 substitutions on average, compared with just 8.0 between any two modern humans, placing their last common mitochondrial ancestor between 550,000 and 690,000 years ago.

The conclusion seemed firm: Neanderthals had not contributed genes to modern humans. But there was a critical limitation researchers acknowledged at the time — mitochondrial DNA is inherited only through the maternal line, cannot detect paternal interbreeding, and represents only about 0.0005 percent of the genome. “No Neanderthal contribution to mtDNA” does not necessarily mean “no Neanderthal DNA at all.”

In 2010, Pääbo’s team — now at the Max Planck Institute for Evolutionary Anthropology in Leipzig — published a draft of the nuclear Neanderthal genome in Science, and it was a different story entirely. Comparing Neanderthal nuclear DNA with the genomes of five living people, they found clear, statistically robust evidence that Neanderthals had interbred with the ancestors of all non-African people: between 1 and 4 percent of the nuclear genome of those with European, Asian, and other non-African ancestry traces directly to Neanderthals. Pääbo’s 2022 Nobel Prize in Physiology or Medicine recognised this work. The 1997 conclusion was not wrong — it was accurate but limited to a single, maternally inherited slice of the genome. Science does not stand still.

Later, higher-quality genomes sharpened the picture. A 2017 genome from a Neanderthal woman who lived about 50,000 years ago in Vindija Cave, Croatia, was more closely related to the Neanderthals who actually mixed with our ancestors, and it let researchers identify 10 to 20 per cent more Neanderthal DNA in living people, including variants affecting cholesterol levels and schizophrenia risk. The share now usually quoted for non-Africans is around 2 per cent.

The timing has narrowed too. A 2024 analysis of more than 300 genomes spanning the past 50,000 years concluded that most Neanderthal gene flow happened during one extended period, between about 50,500 and 43,500 years ago. A 2025 study of the oldest modern human genomes yet sequenced, from roughly 45,000-year-old remains at Ranis in Germany and Zlatý kůň in Czechia, dated their Neanderthal ancestry to the same shared event about 45,000 to 49,000 years ago. That implies the ancestors of everyone outside Africa were still one population at that time.

Neanderthal and Denisovan DNA in Living People

The picture of archaic ancestry did not stop with Neanderthals. David Reich, Professor of Genetics at Harvard Medical School, has spent two decades mapping these encounters; his laboratory maintains the Allen Ancient DNA Resource, a curated collection of the world’s published ancient human DNA that passed 10,000 individuals at the end of 2022, and the findings have repeatedly overturned clean narratives. As modern humans spread east into Asia, they encountered the Denisovans — an archaic group known initially from only a finger bone and two teeth recovered from Denisova Cave in Siberia’s Altai Mountains. Genetically distinct from both Neanderthals and modern humans, Denisovans interbred with the ancestors of people now living in Southeast Asia, the Pacific Islands, and Australia.

According to Reich’s research, Denisovan DNA makes up as much as 5 percent of the ancestry of Melanesians, Indigenous Filipinos, and Aboriginal Australians — the highest archaic ancestry of any living population. A 2016 analysis in Current Biology applied new statistical methods to 257 genomes from 120 populations and found more Denisovan ancestry in South Asians than existing models predicted. It also found both archaic ancestries depleted near genes, especially those active in the testes, suggesting that mixing between populations separated for more than half a million years reduced male fertility.

Then, in 2018, a bone fragment from Denisova Cave yielded a finding that brought these ancient encounters into sharp focus: a teenage girl, nicknamed “Denny,” who had a Neanderthal mother and a Denisovan father — a first-generation hybrid of two archaic human species, preserved across 90,000 years, decoded from a fragment of bone smaller than a fingertip. Our tangled ancestry was not merely a statistical inference. It happened, to specific individuals, in specific places.

For years Denisovans had no face. That changed in 2025. Researchers led by Qiaomei Fu of the Institute of Vertebrate Paleontology and Paleoanthropology in Beijing recovered 95 proteins from a nearly complete skull found at Harbin in north-eastern China, at least 146,000 years old and previously named as a new species, Homo longi. The proteins carried Denisovan-specific variants, and mitochondrial DNA extracted from the hardened plaque on its teeth fell within the Denisovan range. Months earlier, ancient proteins had identified a jawbone from the Penghu Channel off Taiwan as a male Denisovan. Together they show a population spread from Siberia to subtropical Asia, with robust jaws and large teeth.

The clean old story of modern humans migrating out of Africa and simply replacing Neanderthals has given way to a more tangled one. And the inherited DNA is not merely ancestral curiosity — it shapes living biology. Neanderthal variants have been linked to immune function, pain sensitivity, and metabolism; Denisovan DNA in Tibetan populations includes the EPAS1 gene variant enabling survival at high altitude. Archaic DNA is functional, and it is doing things in our bodies right now — a theme that connects directly to how inherited experiences and ancestral environments shape our genes across generations.

The Ghost Population Hidden in West African Genomes

The story of archaic interbreeding reaches back into Africa itself — and involves a population we have never identified. In 2020, geneticists Arun Durvasula and Sriram Sankararaman of UCLA published an analysis in Science Advances that sent a quiet shockwave through the field. Using whole-genome data from West African populations — specifically the Yoruba and Mende — they found clear genetic signatures of interbreeding with an archaic hominin that was not Neanderthal, not Denisovan, and not any fossil species currently known to science.

Four West African populations, including the Yoruba and Mende, derive between 2 and 19 percent of their genetic ancestry from a population that had diverged from the main Homo sapiens lineage before the split of modern humans and Neanderthals — somewhere between 360,000 and one million years ago. Some of the archaic segments are common in these populations, which makes them candidates for having been useful. Who were they? The honest answer is that we do not know. No genome has been recovered, no fossil definitively linked. They are known only through the genetic shadow they cast in living people — a ghost population in the most literal sense.

The 2023 stem study added a nuance: some of what looks “archaic” in African populations may reflect the deep internal structure of ancestral modern human populations rather than a separate lineage, and distinguishing the two remains one of the field’s most contested problems. In future, gene-editing tools like CRISPR may let researchers study the functional effects of these archaic variants directly.

What This Means for Homo naledi and the Fossil Record

Homo naledi is a hominin species discovered in 2013 in the Rising Star Cave system of South Africa, in an excavation led by palaeoanthropologist Lee Berger. Over 1,550 fossil specimens were recovered — the largest single-species fossil collection ever found in Africa. Dated between 236,000 and 335,000 years ago, H. naledi lived alongside early Homo sapiens, with a mosaic of features: a brain well under half the size of ours, but hands and feet surprisingly like our own. Some researchers proposed it might have contributed genes to our lineage.

The weakly structured stem model argues otherwise. Because the populations that gave rise to modern humans were only weakly differentiated — genetically and probably physically similar — the model predicts our true ancestral fossils should look roughly like modern humans. Fossils with dramatically different features, like H. naledi‘s small brain case, more likely represent lineages that diverged much earlier and did not significantly contribute to our gene pool. The profusion of strange-looking hominins in Middle Pleistocene Africa may represent our evolutionary cousins rather than our grandparents. Our true ancestral population was less visually spectacular but far more widespread: interconnected groups of people, spread across a continent, sharing genes and slowly diverging over hundreds of thousands of years.

Ancient Southern Africa: 9,000 Years of Genetic Continuity

The 2023 study did not stand alone. A 2024 study in Nature Ecology and Evolution examined ancient DNA from nine people buried at Oakhurst rockshelter in South Africa and established 9,000 years of genetic continuity in southernmost Africa — Khoe-San people living in the same region, with the same deep genetic profile, for nearly ten millennia. A study published in Nature in December 2025 analysed genomes from 28 ancient southern African individuals dated between 10,200 and 150 years before present. Those older than about 1,400 years carried variation sitting completely outside the range seen in any living person, though some living Khoe-San retain up to 80 per cent of that ancient ancestry. Variants that changed on the branch leading to all Homo sapiens were enriched for genes associated with kidney function, while some that changed only in ancient southern Africans were enriched for protection against ultraviolet light. Genes flowed out of the region more than 8,000 years ago, but outside ancestry flowed in only after about 1,400 years ago.

And the February 2026 Nature Communications study identified 1,376 genes under positive selection in these populations, of which 479 appear specifically associated with forager lifestyles — genes shaped by tens of thousands of years of a particular way of living. Together, these findings say something that should stop us in our tracks: southern Africa is not merely the endpoint of some ancient migration. It is one of the crucibles of what we are. The people who live there today carry, in their DNA, the deepest signatures of human history on Earth. This deep continuity connects to broader questions about how environment shapes gene expression, explored in our article on epigenetics and how your surroundings rewrite the way your genes work.

What Scientists Say

Henn’s co-author Tim Weaver, a fossil expert at UC Davis, has said that where earlier, more complicated models proposed contributions from archaic hominins, this one indicates otherwise.

Even the reassessment of the 1997 mitochondrial work shows how tangled the story is: that study genuinely refuted the idea of Neanderthals as direct lineal ancestors, even though the 2010 nuclear genome later showed that interbreeding did occur, through a different mechanism.

Why This Matters Beyond Academia

Medicine and health. The 2026 Nature Communications study identified over 1.3 million previously unknown genetic variants in Khoe-San populations. Medical genetics has historically been built overwhelmingly on data from European populations — a bias that distorts disease-risk prediction and treatment for billions of people. Understanding African genetic diversity is not merely of historical interest; it is a pressing medical necessity, because every unknown variant in an underrepresented population is a potential clue about disease mechanisms current medicine is missing.

Functional archaic DNA. The Neanderthal and Denisovan DNA in modern genomes is not inert. The EPAS1 variant that helps Tibetan populations survive at altitude — one of the strongest signals of natural selection ever identified in humans — traces to Denisovan ancestry. These inherited variants help explain why different populations have different disease susceptibilities, drug responses, and baseline physiologies.

A more honest story. There is something quietly important about replacing the myth of a single perfect ancestor with the truth of an interconnected network. Humanity did not begin as one isolated group that survived while everything else failed. We began as many groups, across a vast continent, sharing genes and knowledge across long distances for hundreds of thousands of years. The richness of what we are came from that interconnection — not from isolation. And because some inherited archaic variants influence tumour-suppression pathways, the boundary between ancestral variation and disease-relevant mutation is directly relevant to the genetics of cancer and how tumours hijack DNA.

Frequently Asked Questions

Sources

• Ragsdale, Weaver, Henn, Gravel et al. — A Weakly Structured Stem for Human Origins in Africa (Nature, 2023)
• UC Davis / ScienceDaily — DNA Research Rewrote the Origin of Human Species (2026)
• Krings et al. — Neandertal DNA Sequences and the Origin of Modern Humans (Cell, 1997)
• Durvasula & Sankararaman — Ghost Archaic Introgression in African Populations (Science Advances, 2020)
• Schlebusch et al. — Khoe-San Genomes and the Deepest Population Divergence (MBE, 2020)
• A Catalogue of Early-Diverged Khoe-San Genome Variation (Nature Communications, 2026)
• Smithsonian Human Origins Program — Ancient DNA and Neanderthals
• Schlebusch et al. — Southern African ancient genomes estimate modern human divergence to 350,000 to 260,000 years ago, Science 358, 652 (2017)
• Jakobsson et al. — Homo sapiens-specific evolution unveiled by ancient southern African genomes, Nature 650, 156 (2026)
• Gretzinger et al. — 9,000 years of genetic continuity in southernmost Africa demonstrated at Oakhurst rockshelter, Nature Ecology & Evolution 8, 2121 (2024)
• Prüfer et al. — A high-coverage Neandertal genome from Vindija Cave in Croatia, Science 358, 655 (2017)
• Iasi et al. — Neanderthal ancestry through time: insights from genomes of ancient and present-day humans, Science 386, eadq3010 (2024)
• Sümer et al. — Earliest modern human genomes constrain timing of Neanderthal admixture, Nature 638, 711 (2025)
• Sankararaman et al. — The combined landscape of Denisovan and Neanderthal ancestry in present-day humans, Current Biology 26, 1241 (2016)
• Fu et al. — The proteome of the late Middle Pleistocene Harbin individual, Science 389, 704 (2025)
• Fu et al. — Denisovan mitochondrial DNA from dental calculus of the >146,000-year-old Harbin cranium, Cell 188, 3919 (2025)
• Tsutaya et al. — A male Denisovan mandible from Pleistocene Taiwan, Science 388, 176 (2025)
• Hublin et al. — New fossils from Jebel Irhoud, Morocco and the pan-African origin of Homo sapiens, Nature 546, 289 (2017)
• Vidal et al. — Age of the oldest known Homo sapiens from eastern Africa, Nature 601, 579 (2022)
• Mallick et al. — The Allen Ancient DNA Resource (AADR): a curated compendium of ancient human genomes, Scientific Data 11, 182 (2024)