Neanderthals Are Still Inside You: The Two Percent That Never Left

Begin with how solid the ground is, because everything here rests on it. Sequencing the Neanderthal genome ranks among the most significant achievements in all of modern biology, and it is not a claim any fringe would make. The ancient-DNA work behind it won Svante Paabo the 2022 Nobel Prize in Physiology or Medicine, awarded for his discoveries concerning the genomes of extinct hominins and human evolution. The research file's own verdict is blunt: this is one of the best-established findings in evolutionary genetics. So when you read the tiers below, understand that the core of this article is not a theory anyone is still fighting about. It is textbook.

The proof arrived in 2010. Richard Green and colleagues published the first draft sequence of the Neanderthal genome in Science, reconstructed from bones recovered at Vindija Cave in Croatia. It showed that all non-Africans carry between one and four percent Neanderthal ancestry, later refined toward the 1.5 to 2.5 percent figure above, and that the amount looked, at first, roughly equal in Europeans and East Asians. That early equality carried a clue: the interbreeding had to have happened before those two populations split, most likely in the Near East somewhere around fifty to sixty thousand years ago, as the ancestors of everyone outside Africa were passing through.

Since that first rough draft, the genomes have only gotten sharper. Kay Prufer's team sequenced the Altai Neanderthal, roughly 120,000 years old, at fifty-fold coverage in 2014, then the Vindija 33.19 Neanderthal, roughly 52,000 years old, at thirty-fold coverage in 2017. Fabrizio Mafessoni and colleagues added a high-coverage Chagyrskaya Neanderthal, roughly 80,000 years old, in 2020. These are not blurry fragments. They are high-resolution readouts of an extinct human genome, precise enough to line up gene by gene against your own.

The main event has a place and a date. Around fifty to sixty thousand years ago, as anatomically modern humans expanded out of Africa and pushed through the Levant and the Near East, they met Neanderthals who were already living there, and they had children together. That single stretch of encounters accounts for the majority of the Neanderthal ancestry shared by every non-African population walking the Earth today. One region, one narrow window of time, written into billions of living genomes.

It was not the only meeting. Beyond that primary pulse there is clear evidence for at least one additional interbreeding event that fed extra Neanderthal ancestry specifically into East Asian populations, work laid out by Benjamin Vernot and Joshua Akey in 2015. There is also a possible separate, distinct introgression into the ancestors of Melanesians. The picture is not one clean handshake between two species. It is several, spread across tens of thousands of years and thousands of miles.

And sometimes the encounter was startlingly close. The Oase 1 individual, a roughly 40,000-year-old modern human found in Romania, carried between six and nine percent Neanderthal DNA, and it sat in unusually long, unbroken stretches. Inherited DNA gets chopped shorter with every generation, so segments that long mean a Neanderthal ancestor only four to six generations back, a great-great-great-grandparent, give or take. This is Qiaomei Fu's 2015 finding, and it is interbreeding caught almost in the act. Here is the twist that keeps it honest: Oase 1's own lineage led nowhere. It contributed nothing to modern Europeans. It is a dead-end branch on the family tree, proof that the mixing happened without being proof of your particular descent.

In February 2026 a paper in Science added a detail nobody expected: the mixing was lopsided by sex. Alexander Platt, Daniel Harris, and Sarah Tishkoff found that gene flow ran predominantly one direction, between Neanderthal males and modern human females. The fingerprint is on the X chromosome. Neanderthal X chromosomes carry a 62 percent relative excess of modern-human DNA compared with their other chromosomes, while modern humans in turn carry very little Neanderthal ancestry on the X at all. Because females carry two X chromosomes and males carry one, a mating pattern skewed toward Neanderthal fathers and human mothers would let fewer Neanderthal X chromosomes into the human gene pool from the very start. This is one recent paper, not yet weighed by years of follow-up, so it sits at Tier 2, and it does not overturn the older explanation so much as sit alongside it. But it proposes something vivid: not just that two species mixed, but who tended to pair with whom.

Some of what they gave you is useful, and the immune system is the clearest case. Neanderthal-derived versions of the genes TLR1, TLR6, and TLR10, a cluster of Toll-like receptors that form part of your innate immune system, the body's first-line pathogen sensors, are among the best-documented examples of beneficial archaic DNA in living people, work published by Michael Dannemann and colleagues in 2016. They likely shape how your innate immunity responds to microbes. Other immune loci, including some HLA-related signals, have been proposed too, though not every candidate carries equally strong evidence.

Skin and hair are the other recurring theme. Neanderthal-derived variants near the gene BNC2 and across keratin-related pathways are tied to pigmentation, skin, and hair biology in modern humans. It makes a certain evolutionary sense. Neanderthals had spent hundreds of thousands of years adapting to Eurasian light and climate before modern humans ever arrived, and inheriting a few of their skin and hair settings may have been a shortcut worth keeping.

Here is where you have to be careful, because the single most famous story of an archaic gene helping modern humans is not actually Neanderthal. The EPAS1 variant that lets Tibetans thrive in the thin air of high altitude, the textbook case of beneficial archaic DNA, came from Denisovans, the Neanderthals' sister lineage, not from Neanderthals themselves, per Emilia Huerta-Sanchez and colleagues in 2014. Popular articles blur the two constantly. Do not. If you want a genuine Neanderthal gene doing you a favor, point at the immune receptors or the skin genes above, never the high-altitude one.

And the gift is double-edged. Some Neanderthal haplotypes that may once have been advantages now show context-dependent links to inflammatory, metabolic, or infectious-disease traits. The logic is that an adaptation tuned for an Ice Age Eurasian body can read as a liability in a modern one, warm, fed, medicated, and living far longer than any Neanderthal did. Same gene, different world, different verdict.

If interbreeding were the whole story, your genome would be an even blend. It is not, and the unevenness is the tell. In 2014 Sriram Sankararaman and colleagues mapped Neanderthal ancestry across modern genomes and found deserts of introgression, whole regions swept nearly clean of Neanderthal DNA. Three patterns stand out. The X chromosome carries about five times less Neanderthal ancestry than the other chromosomes, exactly what you would expect if hybrid males were less fertile, since the sex chromosomes drive that kind of incompatibility. Genes switched on in the testes are depleted too, more evidence that Neanderthal versions hurt male fertility in hybrids. And genes highly active during brain development are stripped of Neanderthal DNA, hinting that some Neanderthal neurological variants simply did not work in a modern human head.

Those deserts were carved out slowly, by selection. Kelley Harris and Rasmus Nielsen estimated in 2016 that right after the interbreeding, early modern humans may have carried around ten percent Neanderthal ancestry, five times today's level. Over roughly fifty thousand years, weak but relentless purifying selection pared that down to about two percent, quietly deleting the slightly harmful Neanderthal variants generation after generation. The two percent you carry is not what was handed over. It is what survived the edit.

Ivan Juric's team modeled why the purge happened at all, in 2016, and the answer is population size. Neanderthals lived in small, scattered bands, with an effective population roughly forty percent smaller than that of modern humans. In a small population, weakly harmful mutations slip through, because random drift outmuscles selection. Once those same variants crossed into the far larger modern human population, selection could finally see them clearly, and it went to work removing them.

The Y chromosome tells a matching story. Martin Petr and colleagues traced the Neanderthal and Denisovan Y chromosomes in 2020 and found further sex-linked incompatibilities and replacements deep in these archaic lineages. The lesson repeats: some stretches of the genome were open borders for ancient DNA, and others were nearly sealed shut.

Start with a genuine puzzle the field has not fully closed. East Asians carry roughly 15 to 20 percent more Neanderthal ancestry than Europeans do, and there are three live explanations. Maybe East Asia got an extra dose of admixture. Maybe Europeans had theirs diluted, by later mixing with a group called Basal Eurasians who split off before the Neanderthal encounter and were carried into Europe by Neolithic farmers. Or maybe selection simply worked at different strengths in the two populations. The dilution model, from Iosif Lazaridis and colleagues in 2014, is currently the favored answer, but this one is still Tier 2, credible and debated rather than closed.

The clearest way to feel your Neanderthal inheritance is in a doctor's office. Association studies, notably Corinne Simonti and colleagues in 2016, have linked Neanderthal-derived DNA to raised risk for depression and mood disorders, actinic keratosis (sun-related skin damage), hypercoagulation (a tendency to clot), nicotine addiction, and urinary tract dysfunction. Read that list with the tier in mind. Each individual effect is small, a nudge and not a sentence, and the associations are statistical, not destiny. But they are real enough to keep turning up in biobank-scale data.

Then a pandemic put the whole question on the front page. A Neanderthal-derived stretch of chromosome 3, inherited from Vindija-like Neanderthals, turned out to be the single strongest known genetic risk factor for severe COVID-19, with an odds ratio of about 1.6, reported by Hugo Zeberg and Svante Paabo in 2020. It is common in South Asian populations, around 50 percent, moderate in Europeans, around 16 percent, and rare in East Asians, around 4 percent. A gene variant fifty thousand years old was helping decide who ended up on a ventilator in 2020. And the balance sheet is not all bad: a separate Neanderthal haplotype, on chromosome 12, turned out to protect against severe COVID-19 instead.

So is the Neanderthal in you a health asset or a health tax? The honest answer is that no one has settled it. Whether these disease links reflect ancient tradeoffs that were once worth making, a mismatch between an Ice Age genome and a modern lifestyle, plain statistical correlation without direct cause, or some mix of all three, remains an open debate. That uncertainty is not a weakness in the science. It is the field being careful about what a correlation can and cannot prove.

You can now put a rough number on your own inheritance without leaving your kitchen. The consumer testing company 23andMe screens each customer against more than 7,462 known Neanderthal-associated genetic variants and reports the count back as a personal Neanderthal Ancestry Report. By the company's own published figures, the average customer carries about 267 of these variants, with observed counts running from zero all the way up to a reported high of 387. Treat this for exactly what it is: a consumer product's individualized comparison, useful for making the abstract feel personal, not a stand-in for the peer-reviewed population percentages above. The science that says about two percent is the number that has been checked to a decimal. This is the number that has been marketed. Both can be interesting; only one is Tier 1.

The same report goes further, and lighter. It ties specific Neanderthal-derived variants to a grab-bag of everyday traits: whether you sneeze after eating dark chocolate, your sense of direction, how often mosquitoes bite you, your body type, your dandruff, even whether you are naturally strong. Enjoy these for what they are, playful color from a consumer product, not peer-reviewed findings carrying anything like the weight of the immune and skin genes established in the lab. If a spit-tube tells you that your poor sense of direction is the Neanderthal's fault, smile, and keep it in the Tier 3 drawer where it belongs.

First myth: that Neanderthals are the specific ancestors of some modern human race, a genetic dividing line between peoples. The data says the opposite. Everyone whose ancestry lies outside Africa carries a similar small proportion of Neanderthal DNA, and the differences between populations are minor, fractions of a percent. Neanderthal ancestry does not line up with racial categories in any meaningful way. If anything, this DNA is a unifier: it marks the shared out-of-Africa journey of nearly everyone on the planet, not a border between them.

Second myth: the shuffling, dim-witted caveman. The archaeological record does not support it, and has not for a long time. Neanderthals made sophisticated tools, controlled fire, buried their dead, used ochre as pigment, cared for their injured and disabled, and may have made cave art. Their brains were large, averaging around 1,600 cubic centimeters, bigger on average than the modern human brain. And they were no failed experiment. Neanderthals endured across Eurasia for more than three hundred thousand years, far longer than our own species has yet managed. Whatever ended them, it was not stupidity.

There is real debate about how close a Neanderthal mind was to yours, and it deserves to stay a debate. Archaeologists increasingly credit Neanderthals with symbolic behavior, personal ornaments, deliberate burial, and complex toolmaking. The cave-painting claims at certain Spanish sites are important but genuinely contested, because both the dating and the interpretation are disputed. Whether Neanderthal cognition was fully equivalent to ours or subtly different is unresolved, and the genomic deserts around brain-development genes are a quiet argument that at least some real neurological difference existed between the two species.

Could they speak? The clues point toward yes and stop short of proof. Neanderthals had a hyoid bone, the small throat bone that anchors the machinery of speech, and they carried the same FOXP2 gene variant linked to speech and language in us. Computer modeling of their ear-canal structure, by Mercedes Conde-Valverde and colleagues in 2021, suggests hearing tuned to exactly the frequency band that human speech uses. It is a suggestive stack of anatomy. But the full neurocognitive wiring that turns sound into language cannot be read off a skeleton or a gene, so genuine Neanderthal speech stays Tier 3, plausible and unproven.

And there may be more ancestors still hiding. Statistical analyses hint at interbreeding with archaic populations other than Neanderthals and Denisovans, so-called ghost lineages known only from the genetic traces they left, never from a single recovered fossil. These inferences are model-dependent, and they are a separate line of research from the Neanderthal story proper. This wing follows that thread on its own, in the Ghost Populations file next door.