Just how Neanderthal are you? Customers of the genetics testing service 23&Me can now find an answer to that question thanks to their just-released Neanderthal Ancestry Estimator.
All non-Africans tested so far owe some of their genetic diversity to interbreeding between Homo sapiens and Neanderthals. This conclusion comes from comparing the draft genome of Neanderthals to the genomes of contemporary humans from around the world.
A team led by Svante Pääbo at the Max Planck Institute for Evolutionary Anthropology in Leipzig, Germany, reported in 2010 that European and Asian populations owe about 1–4% of their genomes to Neanderthals (see ‘European and Asian genomes have traces of Neanderthal‘). Sub-Saharan African genomes, meanwhile, do not contain these traces of Neanderthal DNA.
Eurasians acquired the DNA after the common ancestor of humans and Neanderthals went their separate ways around a half million years ago, and the ancestors of all non-African Homo sapiens interbred with Neanderthals less than 100,000 years ago, probably somewhere near the Middle East (see ‘Ancient DNA reveals secrets of human history‘).
Pääbo’s team used whole-genome sequences to determine that humans and Neanderthals interbred, and his team and other scientists have developed numerous ways of detecting interbreeding, many using whole-genome data.
Most people don’t have their whole genomes sequenced (at least not yet), but tens of thousands of people have gotten genome scans from consumer genetic-testing firms like 23&Me and Iceland-based deCODE. These scans contain the sequence of hundreds of thousands of individual letters in the human genome called single-nucleotide polymorphisms, or SNPs (pronounced ‘snips’), which tend to vary from person to person.
In theory, these scans contain enough information to estimate an individual’s Neanderthal heritage, but one of the scientists involved in the Neanderthal genome project, Harvard Medical School’s David Reich, told me in 2010 that it would be a challenge to use SNP data for this purpose.
Now his collaborator on the genome and current 23&Me employee Eric Durand has come up with a way to estimate human–Neanderthal interbreeding using 23&Me genome scans (he describes it in this white paper).
This is a massive over-simplification, but Durand’s method distils the several hundred thousand SNPs into two numbers, called principle components, which can be graphed on x- and y-axes. The method then compares the distance of an individual’s genome to that of Neanderthals. They define 0% Neanderthal interbreeding based on the coordinates of 246 Africans, so the closer your genome is to that of the Neanderthals and the further from that of the Africans, the more Neanderthal ancestry it contains.
Luke Jostins, a graduate student at the Wellcome Trust Sanger Institute near Cambridge, UK, who blogs at Genomes Unzipped, kindly shared the results of his Neanderthal test. It revealed that he owes 3.1% of his genetic diversity to interbreeding with Neanderthals (23&Me doesn’t include error bars with their analysis, and Jostins’s Genomes Unzipped colleague Daniel MacArthur noted that his Neanderthal heritage varied depending on the gene chip used). Here’s a screen shot:
Jostin’s 3.1% figure is slightly above average for people of European ancestry. “I am a strong believer in hybrid vigour, so I am going to add the above-average non-sapiens score to the ‘positive’ column of my overall genetic health,” he writes in an e-mail. He has shared some of the other insights that he has gained from his genome here.
The 23&Me test does not identify which SNPs its customers acquired from Neanderthals, but with some urging, Jostins speculated on what traits he might have inherited. “Knowing my luck, I probably have a bunch of Neanderthal olfactory receptors: ready to smell a bunch of long-extinct European forest flowers. Oh, I love winters and hate summers! That’ll be it, three one-hundredths of my genome is always longing for the glacial Pleistocene,” he says.
Speculation aside, identifying which genes and even traits humans acquired from Neanderthals is a hot area of research. Pääbo’s team reported a number of regions in the human genome that may have been inherited from Neanderthals, and Damian Labuda, of the University of Montreal, Canada, and his colleagues recently identified a version of a gene called dystrophin that humans almost certainly inherited from Neanderthals. There will be many more reports like this to come.
Even more exciting is the possibility that humans have genes from Neanderthals that offered our ancestors useful traits. Peter Parham’s team at Stanford University in California reported this summer that humans may have gotten immune genes from both Neanderthals and Denisovans, another ancient hominin population that lived in Asia. These genes may have helped humans leaving Africa fend off pathogens in Europe and Asia, but some scientists have criticized this conclusion as premature.
23&Me’s Neanderthal Ancestry Estimator may be genomic navel-gazing of the most trivial (and most interesting) kind. But there are plenty of good reasons to look for signs of interbreeding with Neanderthals and Denisovans in as many humans from around the world as possible.
Mark Stoneking, a colleague of Pääbo’s at the Max Planck Institute, measured Denisovan ancestry in numerous individuals from all over Asia to trace Homo sapiens migration routes through Southeast Asia and Oceania (see ‘First Aboriginal genome sequenced’).
And one paradox of the Neanderthal interbreeding story has always been that there is no sign that Europeans, whose ancestors may have overlapped with Neanderthals for several thousand years in places, show no more signs of admixture than Chinese, Papua New Guineans or other non-African groups (See ‘Who were Europe’s first humans?‘).
There are plenty of good explanations for this (humans and Neanderthals interbred only in the Middle East, or perhaps the humans who interbred with Neanderthals in Europe went extinct). But tests like 23&Me’s have the potential to turn up individuals whose Neanderthal ancestry far exceeds 4%, and these people could be the descendants of another series of trysts between humans and Neanderthals.
Update 16 December
Some people have posted their results to Twitter under the hashtag #Iamthe1to4percent.
Update 19 December
Harvard Medical School’s David Reich writes via e-mail: “This looks like a serious calculation.” However he questions the precision of the estimate. The range of Neanderthal ancestry among non-Africans that 23&Me calculates probably represents statistical noise and not true individual variation, Reich says.
“The only population for which the Neandertal calculator seems to be giving meaningful estimates seem to be African Americans, and this is entirely mediated through their African ancestry. In other words, the test is providing no information beyond the estimate of a person’s African ancestry, which is due to much more recent phenomena,” he adds.
Svante Pääbo tells Nature in an e-mail that his team considered patenting a Neanderthal ancestry test before they published the genome, so as “to have some influence in the future on how individual Neandertal ancestry was implemented and presented to the public.” They decided against the idea.
Pääbo thinks the 23&Me test is “OK” as an overview, but he thinks a more interesting test would identify individual segments of the genome that individuals have inherited from Neanderthals. “This will be possible in the future,” he says.
Image of Neanderthal reconstruction courtesy Wikimedia Commons