By Michael Marshall
A visualisation of relationships between ancestors and descendants in the genealogy of modern and ancient genomesWohns et al. (2022)
Meet your relatives. A family tree of humanity has been constructed using genetic data from thousands of modern and prehistoric people. The tree gives a view of 2 million years of prehistory and evolution.
“Humans are all ultimately related to each other,” says Gil McVean at the University of Oxford. “What I’ve long wanted to do is to be able to represent the totality of what we can learn about human history through this genealogy.”
The new family tree suggests that our earliest roots were in north-east Africa. It also offers clues that people reached Papua New Guinea and the Americas tens of thousands of years earlier than the archaeological record implies, hinting at early migrations that haven’t yet been discovered. However, both these ideas would need to be confirmed by archaeologists.Advertisement
Join us for a mind-blowing festival of ideas and experiences. New Scientist Live is going hybrid, with a live in-person event in Manchester, UK, that you can also enjoy from the comfort of your own home, from 12 to 14 March 2022. Find out more.
Geneticists have been reading people’s entire genomes for the past two decades. McVean and his colleagues compiled 3609 complete genomes, almost all of which belonged to our species, Homo sapiens, except for three Neanderthals and one from the Denisovan group, which may be a subspecies of H. sapiens or a separate species.
Putting them together into a tree was challenging. “The different data sets have been produced over time, using different technologies, analysed in different ways,” says McVean.
The team focused on bits of DNA that vary from person to person. They identified 6,412,717 variants and tried to figure out when and where each one arose. To do this, they also looked at an additional 3589 samples of ancient DNA that weren’t good enough to include in the tree, but did shed light on when the variants emerged.
Variants that emerged before 72,000 years ago were most common in north-east Africa, and the oldest 100 variants were also from there, specifically in what is now Sudan. Those oldest variants are about 2 million years old, so they long predate our species, which emerged around 300,000 years ago. Instead, the variants date to the earliest members of our genus, Homo.
The simplistic interpretation of this is that humanity first evolved in that region, but it is likely that subsequent migrations have interfered with the data. “I would definitely not take the naive and immediate answer,” says Jennifer Raff at the University of Kansas.
The earliest H. sapiens fossils are from the north and east of Africa, but few have been discovered, so we don’t know our species’ early range with any certainty. The oldest known specimens are from Jebel Irhoud in Morocco, in north Africa, and are perhaps 315,000 years old. The next oldest are those from Omo-Kibish in Ethiopia, in the east. They were thought to be 197,000 years old, but a paper published in January presented evidence that they are more like 233,000 years old.
Many anthropologists now think there were multiple populations spread across Africa, which were sometimes separated and sometimes interbred. If that is correct, humanity doesn’t have a central origin point. “Our findings are certainly perfectly compatible with that,” says McVean. “There’s a lot of very deep lineages within Africa, which are suggestive of that notion of there being multiple source populations, very deeply diverged, representing really ancient splits.”
In line with this, a second study published this week obtained ancient DNA from six sub-Saharan African people who lived within the past 18,000 years. They carried DNA from three distinct lineages that originated in the distant past, from eastern, central and southern Africa. These groups began interbreeding more around 50,000 years ago, but by 20,000 years ago this largely stopped.
The new genealogy also contains hints of early journeys. It suggests that people were living in Papua New Guinea 140,000 years ago, almost 100,000 years before the earliest documented inhabitants. Similarly, it indicates that people were in the Americas 56,000 years ago, despite many archaeologists having settled on 18,000 years ago as the earliest entrance.
The idea of people in the Americas earlier than this is controversial because, prior to that, great ice sheets covered the northern regions, blocking migration. Nevertheless, a study from September 2021 described footprints from White Sands National Park in New Mexico, which suggest humans were in the Americas between 23,000 and 21,000 years ago. There is also disputed evidence of humans living in Chiquihuite cave in Mexico as much as 33,000 years ago. But 56,000 years ago is still a big reach.
“I think there are three possible explanations,” says McVean. “One is, we’re wrong.” The second is that people really were in these places very early.
The third option is a more complex scenario. The first people to live in the Americas came from eastern Asia, and it may be that the population from which they came has died out in Asia. This would mean the oldest American-looking genetic variants are actually from people who lived in Asia – but the only living people with those variants today are in America, throwing off the analysis. A similar story could have played out for Papua New Guinea.
“It’s very common in our genetic data that there are ancient lineages which don’t persist throughout time,” says Raff. “That’s completely plausible.”
Journal reference: Science, DOI: 10.1126/science.abi8264
Sign up to Our Human Story, a free monthly newsletter on the revolution in archaeology and human evolution
More on these topics: More