20 September 2012
The complete genome of a Denisovan – an ancient human that lived at the same time as Neanderthals – has been sequenced using a tiny finger bone and a new, cutting-edge technique offering unprecedented accuracy.
As a result of the new technique and an extremely well preserved bone, the sequence is more precise than the famous culmination of the Human Genome Project published in 2003 – and on par with what is achievable when sequencing modern humans today.
“There is actually today no difference in what we can learn genetically about a person that lived 50,000 years ago and from a person today, provided that we have well-enough preserved bones,” Svante Pääbo, paper co-author and a geneticist from the Max Planck Institute for Evolutionary Anthropology in Germany, said in a press conference.
Papuans have 6% Denisovan DNA
Using the DNA, the researchers dated the Denisovan remains, which appear to have belonged to a female child, who died between 74,000 and 82,000 years ago. The findings, which were the result of a massive international collaboration involving more than 30 scientists, are published in the journal Science.
The new research adds evidence to the recently challenged theory that our species, Homo sapiens, interbred with Neanderthals as well as, it now seems, Denisovans.
While Neanderthals contributed to the DNA of all modern humans excluding indigenous Africans, Denisovan DNA can only be found in people indigenous to Papua New Guinea, Fiji, Australia and other parts of Melanesia. In fact, the researchers estimate that up to 6% of the genomes of modern-day Papuans derive from Denisovans.
Different human groups, not species
The new research also confirmed earlier studies that showed the Denisovan individual carried the same genes that are today associated with having dark skin, brown hair and brown eyes.
“However, since we have access to only a single Denisovan individual, we expect that only a subset of these would have been shared among all Denisovans,” the researchers wrote in the paper. In other words, the little girl may simply have had dark hair and eyes, traits that may not have been shared by all Denisovans.
Pääbo said that he avoids referring to archaic humans, such as Denisovans, as separate species, during a press conferences
“We say it's a different group of humans,” he said. “[Denisovans] are about as different from Neanderthals … as people in Africa would be from people in Scandinavia. I wouldn't call the Neanderthals a different species from [modern] humans, either, actually,” he added.
Denisovan DNA is present in modern people in the Philippines, Indonesia, Fiji, but the biggest concentration is right here.
Dr Jonica Newby
So this is the one that interests me here. This group of Australians contain the highest percentage, along with the New Guineans, of Denisovan DNA to this day?
Dr Mark Stoneking
That's right. The amount of Denisovan ancestry in native Australians and New Guineans we estimate at somewhere around four to six percent.
Dr Jonica Newby
Four to six percent. Wow.
Which means any Australian with an Indigenous ancestor probably has Denisovan as well as Neanderthal genes. So we have a picture, now, of 50,000 years ago - Neanderthals in the west and Denisovans in the east - with the most likely place that Denisovans bred with Homo sapiens being down here in eastern Indonesia.
Revolutionary new sequencing technique
While tiny, the Denisovan fingertip bone discovery was an enormous stroke of luck. The bone turned out to contain 70% ‘endogenous’ DNA – DNA belonging to the owner of the finger, and not to bacteria and other contaminants. That is compared to an average of about 1–5% endogenous DNA in other ancient human remains.
This, combined with the innovative technique that allows geneticists to sequence from single-stranded DNA, meant the researchers could map the Denisovan genome with an extremely high ‘coverage’, or accuracy – better than anything achieved with Neanderthals, despite the much larger volume of Neanderthal remains available.
‘Coverage’ in genome sequencing refers to the average number of times each DNA base is sequenced – the higher the number, the higher the overall accuracy. The Denisovan genome was sequenced with 30-fold coverage. By comparison, when the first human genome was sequenced in 2003, it had a 10-fold coverage. For the first draft of the Neanderthal genome presented in 2010, each DNA base was sequenced just once on average.
“This new sequencing technology brings tears to my eyes,” said David Lambert, a geneticist and evolutionary biologist from Griffith University in Queensland, Australia. Lambert wasn’t involved with the Denisovan research, but is working on sequencing the genomes of the oldest human remains found in Australia. “Technology like this is very important to us. Let’s hope it’ll be a major assistance,” he said.
Genomes are best sequenced from whole, double-stranded DNA. However, DNA found in ancient remains has usually degraded over time to fragmented strands. The new technique works by manipulating these ancient DNA fragments by attaching artificial ‘adapters’ from which the fragments can be copied or ‘amplified’.
“By doing this, we kind of succeeded to develop a more efficient way of extracting information from the few DNA fragments preserved in the bone,” explained another paper co-author Matthias Meyer, also from the Max Planck Institute.
Denisovan population declined as Homo sapiens numbers grew
Using this new technique, the researchers were also able to track the genetic diversity of the Denisovans and therefore any historic fluctuations in population numbers. A striking find was that around the same time modern-day human populations spiked, between 125,000 and 150,000 years ago, Denisovan population saw a “drastic decline”, researchers said in the paper.
Based on the findings of this population "bottleneck", “it is tempting to think that there was an interaction between the two [human groups],” said Lambert.
The research paper also pinpointed certain genes that were fixed in modern humans after the divergence from our ancient relatives between 170,000 and 700,000 years ago.
“It is very interesting that a lot of those genes are important in terms of things like brain function and the the nervous system,” said Lambert. “Things like that are really starting to give us a much better idea of what it means to be Homo sapien.”
The scant remains of the Denisovan girl were discovered in the remote Denisova Cave in the Altai Mountains, southern Siberia, in 2008. Originally thought to belong to a Neanderthal variant, the remains turned out to be that of a previously unknown type of human, which was named after the discovery site.
Denisovans represent the first time an extinct group of humans has been defined as such from DNA sequencing alone, and not from the form and structure of the bones. The researchers said they hope to sequence the Neanderthal genome to a similar level of accuracy and coverage using the new technique within a year.
“I would also not be totally surprised if in the future one finds other groups of humans in addition to Neanderthals and Denisovans out there, particularly in Asia,” said Pääbo.
The discovery that a second branch of the human family shared Asia with our ancestors and the Neanderthals was a real shock, but the Denisovans have continued to surprise many. All we have of them is a bit of a finger and some molars, but those few fragments have yielded a wealth of DNA, and with it the knowledge that the Denisovans interbred with the ancestors of some modern human populations. Now, with the help of a new approach to sequencing ancient DNA, we actually know more about the Denisovans' genome than we do about Neanderthals'. In the process, we've discovered some of the changes that are distinct to modern humans.
We've been attempting to sequence Neanderthal DNA for a while, but progress has been limited by the fact that less than five percent of the DNA we can get out of Neanderthal samples actually came from that individual—most of the DNA is bacterial, and there have been problems with contamination by modern human DNA. But the Denisovan samples have been completely different, as roughly 70 percent of the DNA obtained from them actually comes from the bones.
The problem now is that we simply have so few samples that, even with such a fantastic yield, we were going to run out of DNA. So the people behind the genome effort (led by Svante Pääbo of the Max Planck Institute) came up with a new sequencing technique. Instead of working with double-stranded pieces of DNA, they separated the two strands and amplified and sequenced them separately. Now, even if one of the strands was damaged, it was possible to obtain sequence from the other. With the new technique, they were able to sequence each individual base an average of 30 times. Over 99.9 of the bases in the genome were sequenced at least once.
That level of coverage is 20 times what we have available for Neanderthals. So, even though we have a host of Neanderthal skeletons and artifacts, we know more about the Denisovan genome.
What has it told us? For one, based on the genome, the Denisovans had dark skin, eyes, and hair. It appears modern humans and Denisovans separated about 800,000 years ago. But the lineages came back in contact via the ancestors of modern-day Papuans. A full six percent of the Papuan genome appears to be derived from interbreeding with the Denisovans. And, intriguingly, the amount of Denisovan DNA was a bit lower on the X chromosome. One possible explanation for this if is male Denisovans did most of the mating with modern humans.
The Denisovan and modern human lineages should have been picking up mutations at similar rates since the split. But, since this particular Denisovan died a while back, its genome should have ended up with fewer mutations than a modern human genome. By calculating this deficiency and then working back from the split with chimps, the authors estimate that the owner of the finger and molars died about 75,000 years ago.
Even though the sequence came from a Denisovan, it can actually tell us a lot about ourselves. With high quality sequences of humans, chimps, and Denisovans, it was now possible to identify the changes that are common to the human lineage (found in both us and Denisovans) along with those that are unique to the modern human lineage (present in us, missing in chimps and Denisovans).
For example, one of our chromosomes is a fusion between two separate ones found in chimps. We can identify the sequences that were fused in the human genome, which makes a powerful argument for common descent (much more on that here). Since we know precisely where in the genome the fusion occurred, the researchers searched for that location in the Denisovan genome. It looked like the one in modern humans, indicating that the chromosomes fused deeper into our past.
Overall, the authors found 111,812 single DNA base differences and 9,499 insertions and deletions that are distinct to modern humans. But very few of these were in areas known to affect the expression or structure of genes—just a bit over 100. Most of the rest of the changes were either in DNA that appears to be non-functional, or in areas that tolerate a lot of variability. A number of the changes that did affect genes altered ones that control the development of the nervous system. Others were found in genes that cause human diseases, many of them affecting the skin and eyes.
Overall, the changes within genes are relatively few, so we may have to look outside of the areas that code for proteins to find out more. But, as with the Neanderthal sequence, it may be that the differences between us and some of our closest relatives are going to be subtle, and very difficult to spot. And it's possible that there will be fewer than some of us modern humans would like to believe.