Note: This article is excerpted from the book "Neanderthals" (Published by Zhejiang Education Press , produced by Houlang, December 2018). Author Svante Pääbo, winner of the 2022 Nobel Prize in Physiology or Medicine , was born in Stockholm, Sweden in 1955, and obtained a doctorate degree from Uppsala University in 1986, and later engaged in postdoctoral research at the University of Zurich, Switzerland and the University of California, Berkeley in the United States. In 1990, he became a professor at the University of Munich, Germany, and founded the Max Planck Institute for Evolutionary Anthropology in Leipzig, Germany in 1999. He is still active in the institute. He is also an academician of the National Academy of Sciences in the United States, , Sweden, and other countries. His father, Suen Bergstrom, was the Nobel Prize in Physiology or Medicine in 1982.
In May 2009, we began sequencing the genomes of 5 modern humans. These original DNAs were not contaminated and chemically damaged by bacterial contamination and chemical damage like Neanderthal samples, so we obtained from each sample about 5 times the number of DNA sequence . A year or two ago, we could not imagine sequencing these genomes in Leipzig . However, as soon as companies such as 454 and Yimanda are introduced to the market, a small research team like us can complete the whole genome sequencing task for several people in just a few weeks.
Ed used the method he described in Dubrovnik to estimate how long ago the genomes of these 5 modern humans had common ancestors with the human reference genome. He discovered that Europeans, Papuans and Chinese had common ancestors with the human reference genome more than 500,000 years ago. If San from South Africa was joined, the genealogy separated nearly 700,000 years ago. The Sanghe people (and related populations) are separated from other Africans, other people in the world, the longest time. Therefore, the common ancestor of Neanderthal and modern human genomes was about 830,000 years ago, only 130,000 years earlier than the common ancestor of Sanghe and human reference genomes. So the Neanderthals are indeed different from us, but the difference is not very big.
We must be careful with such calculations because they only provide a chronological value of a common ancestor, which seems to be the same for the entire genome. However, the genome is not a whole genetic unit. This means that each part of each individual genome has its own history, so each part and other genomes also have different common ancestors. This is because everyone carries two copies of each chromosome, one of which is passed to the child independently. Therefore, each chromosome has a unique historical pattern, or its own genealogy. In addition, when egg cells and sperm form, each pair of chromosomes exchanges fragments with each other in complex molecular dances, a method also known as recombination. Therefore, not only does each chromosome in the population have its own genealogy, but every fragment of each chromosome is also present. Therefore, the age of common ancestors calculated by Ed using the human reference genome, whether they were shared with Neanderthals 830,000 years ago or shared with Sanghe people 700,000 years ago, these numbers represent only the average of all parts of the genome.
In fact, when we compare the DNA regions of two modern humans to each other, it is easy to find that some regions have shared ancestors tens of thousands of years ago, while some regions have shared ancestors 1.5 million years ago. This result will also be found when we compare the modern human genome to the Neanderthal genome. If someone could walk on one of my chromosomes and compare it to the chromosomes of Neanderthals and those of readers who are reading this book of Genes of Life, then Chromosome ramblers will find that sometimes I am more similar to Neanderthals, sometimes readers more similar to Neanderthals, and sometimes readers more similar to me. Ed’s average only means that when comparing the readers, my genomes with Neanderthals, there are more and more similar genome regions between the readers and me.
83 Ten thousand years is just the average time that modern human DNA sequences and the DNA sequences carried by Neanderthal fossil have common origins. It is very important to realize this. These DNA sequences first existed in a population, and later the descendants of this population became ancestors of Neanderthals and modern humans. But this is not the time when modern humans and Neanderthals separated from each other, and separation must have happened later. Because we trace the DNA sequence history of modern humans and the DNA sequence history of Neanderthals, we found that these two lineages will first enter the last ancestral group of modern humans and Neanderthals (the two populations that were separated later originated from this), and then enter the mutation of this ancestral group at that time. So 830,000 years is a compound time, which includes the time when Neanderthals and modern humans become different populations, and also the time when their common ancestral populations are genetically mutated.
For us, the population of ancestors is still mysterious, although we think it once lived in Africa, and some of its descendants later left there and became the ancestors of Neanderthals, while those who remained became the ancestors of today's humans. Using DNA sequences to estimate the separation time between these two populations is tricky and much harder than using DNA sequences to estimate the time for their common ancestors. For example, if there are many variations between the ancestral groups of Neanderthals and modern humans, if we accumulate more differences in DNA sequences in the ancestral group than after Neanderthals and modern humans part ways, it means that the separation time of the two groups is relatively new. We can estimate the evolution time of different genome fragments to infer the differences and roughly estimate the level of variation in ancestral populations. To estimate the separation time of populations, we also need to know the generation time, that is, the average age of each individual’s offspring. But we obviously know nothing about this time. We did everything we could to take all these uncertainties into account and concluded that the two populations separated sometime between 440,000 and 270,000 years ago. Of course, we may still underestimate a lot of uncertainty. However, the ancestors of today's human beings and the ancestors of Neanderthals had parted ways at least 300,000 years ago.
After estimating the differences between Neanderthals and modern humans, we need to return to the original question: what happened between modern human ancestors left Africa and met long-lost Neanderthal "cousins" in Europe. To understand if there is a gene exchange between modern humans and Neanderthals, Ed compared the genomes of those five modern humans with those of chimpanzees, and David and Nick repeated their own analysis. I believe the result is reliable this time, and the special similarities between Neanderthals and Europeans and Chinese whom I secretly doubted before should disappear.
On July 28, I received two long emails from David and Nick. This is a good witness to David’s passion for science, as his wife Eugenie just gave birth to their first child on July 14, and he is still working on analyzing the work. Nick performed 10 possible pairwise comparisons of 5 modern human genomes. In each case, he identified SNP, which differed between the same chromosome between the two individuals. He found about 200,000 such differences between either pair of genomes. So many SNPs are enough to make us accurate as to which Neanderthals are closer.
Nick found that the matching value between Neanderthals and Sanhe people was 49.9%, and the matching value with the Papuan sample was 50.1%. This is all expected, after all, Neanderthals have never appeared in Africa, so compared to others, Neanderthals will not have a closer relationship with Africans. He found that the Neanderthals matched with the French was 52.4%. We have a lot of data now, so the uncertainty of these values is only 0.4%. It is therefore obvious that the French are more similar to Neanderthals than the genome of the Sanghe people. When he compared the SNPs of French and Yoruba people, he found that the corresponding match value was 52.5%.When the SNPs of Chinese and Sanghe people and Chinese and Yoruba people were adopted, the corresponding matching values were 52.6% and 52.7% respectively; when the SNPs of Papuans and those two Africans were adopted, the matching values were 51.9% and 52.1%, respectively. When using the SNP between French, Chinese and Papuans for analysis, the matching values obtained vary between 49.8% and 50.6%. Therefore, in all comparisons that do not involve Africans, the match values are around 50%. But whenever Africans are compared with those outside Africa, Neanderthals match people outside Africa, 2% more SNPs matched by Neanderthals and Africans. So wherever they have lived, Neanderthals do have small and obvious genetic contributions to people outside Africa.
I read those two emails once and then read them again. This time I read it very carefully, trying to find any analytical problems, but I got nothing. I leaned against the office chair, looking blankly at the messy desk, filled with papers and notes from the past few years, stacked up one by one. The results David and Nick were shown on the computer screen, staring at me straight. This time it was not due to some technical error, but the Neanderthals really contributed DNA to modern humans. This is really super cool. This is the answer I have dreamed of for the past 25 years. The question about human origin has been controversial for decades, and we now have solid evidence to answer this question, and the answer is unexpected. Not all genomic information of today’s humans can be traced back to the recent ancestors of Africa, which contradicts the rigorous “out of Africa” hypothesis, and my mentor Alan Wilson was once one of the main builders of “out of Africa”. This evidence also goes against the theory I have long believed in. Neanderthals were not completely extinct, and their DNA still exists in today’s humans.
I stared at the table blankly, realizing that the unexpected results of our results were not only contradictory to the assumption of “get out of Africa”, but also because it also does not support the “multi-regional origin” hypothesis. Contrary to the prediction of the "multi-regional origin" hypothesis, we not only found Neanderthal genetic contributions among Europeans, but also saw corresponding evidence among Chinese and Papua New Guinea . How is this possible? I began to organize the table absent-mindedly. At first, the movement was slow, but it became more and more exciting. I threw away the project materials I had accumulated over the years, and dust flew between the stacks of books on the table. I need to start a new chapter. I need a clean table.
sorting can sometimes help you think. After the cleaning, I drew arrows on the map, visually indicating the possible paths of modern humans leaving Africa and meeting Neanderthals in Europe. I can imagine them having kids with Neanderthals, and then these kids are integrated into modern humans, but I can’t figure out how Neanderthal DNA has reached East Asia. This is likely to be the subsequent migration of modern humans and then bringing Neanderthal DNA into China. If this is true, then the similarity between Chinese and Neanderthals should be lower than that between Europeans and Neanderthals. But that's not the case. Then I realized: According to the arrow I imagined, modern humans passed through Middle East after they walked out of Africa! This should be the first place where modern humans meet Neanderthals. If these modern humans crossed with Neanderthals and then became the ancestors of all people outside Africa today, then the amount of Neanderthal DNA carried by people outside Africa should be the same (see picture below). This situation should be possible. But experience tells me that sometimes my intuition happens to be wrong. Fortunately, I know that if I'm wrong, people like Nick, David, and Monty who validate theories with mathematical thinking will correct me.
Neanderthal hybridization plan. If early modern humans who left Africa crossed with Neanderthals and continued to reproduce outside Africa, they would carry Neanderthal DNA to areas where Neanderthals had never lived. For example, about 2% of Chinese DNA comes from Neanderthals.
We discussed the results of David and Nick during Friday meetings and during intense league calls. Now some of us believe that Neanderthals have hybrids with modern humans, but others are still reluctant to believe it, and they try to point out where David and Nick's analysis might be wrong. I realized that it would be even harder to convince everyone in the alliance that these results were so difficult, especially since many paleontologists have not found evidence of modern human hybridization with Neanderthals in the fossil record. This includes many people who are highly respected in the field of Paleologies, such as Chris Stringer at the Museum of Natural History in London and Richard Klein at Stanford University in California. Although I think these paleontologists are very cautious in interpreting fossil records, they may still be influenced by previous genetic results during the study. Many research teams, including us, have shown that all genetic variations in humans today are recent genetic variations from Africa. We showed in that 1997 paper that Neanderthals did not contribute any mitochondrial DNA to today's humans. The paper has also had a profound impact. Although paleontologists such as Milford Wolpov of the University of Michigan and Eric Trincos of the University of Washington- St. Louis University of have found evidence of hybridization in fossils, and some geneticists have also tried to point out that genetic variation in modern humans may have come from Neanderthals, these arguments are not enough to shake common understanding in academic circles. At least they didn't cause a strong impact on me. Previously, there was no need to use Neanderthal genetic contributions to explain the morphological or genetic variation patterns of humans in today's world. But now it is different from before, we can directly study it through the Neanderthal genome. We have seen genetic contributions, although very little.
However, I think more evidence is needed to make the world convince our results. Science is not just an objective and impartial search for truths that scientists cannot imagine and cannot be broken. In fact, scientific research is a social work, in which dogma advocated by high-powered people and influential scholars often determines scientific "common sense". To break this common sense, we must do more analysis of the Neanderthal genome, unlike David and Nick's method of calculating SNP alleles. If more independent evidence also suggests that there is indeed gene flow from Neanderthals to modern humans, it will be easier to convince most of the world. Therefore, finding other feasible analyses became a new constant theme for our weekly conference call.
is a bit unexpected, and the feasible advice comes from outside our alliance. At the Cold Spring Harbor Conference in May 2009, David met Rasmus Nielsen. Rasmus is a population geneticist of Danish and completed his doctoral thesis with Monty Slatkin in 1998. He is now a professor of population genetics at the University of California, Berkeley. Rasmus told David that he and his postdoctoral fellow Zhai Weiwei searched the genomes of today's humans by region and found that the mutation outside Africa was larger than that inside Africa. Although this is very likely, this pattern is often beyond our imagination, as small branches separated from larger populations generally contain only a portion of the variation in the ancestral population. If such an area is found, there are many explanations, but there is one that may be very interesting to us. After being separated from modern humans, Neanderthals lived independently outside Africa for hundreds of thousands of years, and must have accumulated genetic mutations that are different from modern humans. If they subsequently contribute genomic fragments to people outside Africa, it is very likely that these genomic regions will be found using Rasmus' method, because the genetic pattern of these regions happens to be more variants outside Africa than within Africa. Using the Neanderthal genome at hand, we can now check whether a portion of the genes from these regions are from Neanderthals, because DNA sequences outside of Africa are closer to Neanderthals in the regions studied by Rasmus.In June 2009, I invited Rasmus and Weiwei to join the Neanderthal Genome Analysis Alliance.
Rasmus focuses on unusual European regions that were separated from Africans very early. He discovered a total of 17 such areas. Ed sent 15 Neanderthal DNA sequences located in these regions to Rasmus. The latter replied in July, and the results were surprising. Of these 15 regions, 13 contain Neanderthal mutations that exist in Europeans but not in Africans. Later, Rasmus further refined the analysis results and focused mainly on 12 regions of the DNA sequence with more than 100,000 nucleotide . He found that in 10 of these areas, the mutations carried by Neanderthals can still be found in today’s Europeans. This is indeed an amazing result! I can't imagine other possibilities except for those who interpreted as genes being transmitted from Neanderthals to Africa. Although this is just a qualitative result that scientists often talk about, and it is impossible to calculate the DNA contribution of Neanderthals to Europeans or Asians, it vividly proves the existence of genetic contributions. Although this method is different from quantitative analysis, it independently verifies David and Nick's methods and draws the same conclusion.
We continue to think about other methods to detect gene flow. David came up with a brilliant idea as always. His idea is simple: some regions of the human genome today are similar to those of Neanderthals because there are few mutations, which may be due to low mutation rates, or the individual death due to mutations. If one of my genome regions is similar to Neanderthals for the same reason, then this region of mine may also be similar to other humans today, as it rarely changes. But if my genomic region is similar to Neanderthals because my ancestors inherited the region from Neanderthals, there is no reason for this genomic region to be similar to other people. In fact, since the evolutionary history of Neanderthals is different from that of modern humans, this area of mine should be more different from that of others.
David set out to use these ideas for practical analysis. He divided the European parts of the human reference genome into fragments, and then compared the fragments with the Neanderthal genome and other European genomes (i.e. Craig Venter's genome) to find out the number of differences. He found that, in general, the closer the European fragments of the human reference genome are to Neanderthals, the closer they are to Craig's genome. This suggests that the mutations accumulated in these fragments determine the differences between the European fragments of the human reference genome and the Neanderthal and Craig genomes. But he found that in European fragments that were very similar to Neanderthals, the more similar these European gene fragments were to Neanderthals, the greater the difference between them and Craig’s genome. Although I have believed that gene flows occur based on other analyses, when David visited our lab in December 2009 and presented these results, I was convinced that the world could be convinced that the fragments of Neanderthal DNA remain in today’s humans. No matter how we study these data, we have all come up with the same results.
Now we can turn our attention to studying the ways and when and where modern humans interact closely with Neanderthals. The first question is, the direction of gene flow: is it that modern humans contributes DNA to Neanderthals, or does Neanderthals contribute DNA to modern humans, or both? Although some people may think that when two human groups meet, genes flow parallelly in both directions, there are few such examples in real life, and one side often occupies a social dominance. One common pattern: dominant males and non-dominant females mated and produced offspring, while the child remained in the non-dominant group where the mother was. Therefore, gene flow will flow from socially dominant groups to non-dominant groups. The obvious examples are the white slave owners of in the southern United States and the British colonialists in Africa and India.
We tend to think that modern humans are stronger than Neanderthals because Neanderthals eventually disappear. But our data suggest that gene flows from Neanderthals to modern humans. For example, David’s latest results show that the DNA regions that Europeans are very similar to Neanderthals are very different from those in other Europeans. This means that sometime before entering the current European gene bank , these regions and those of other Europeans have begun to accumulate differences separately. Now assume that this happens in Neanderthals as well. If the contribution is in another direction (from modern humans to Neanderthals), then these regions will be just general regions of the European genome, and the number of differences should also be the average of other European genomes. Based on these analyses and other reasons, we believe that all (or almost all) of the gene flows are from Neanderthals to modern humans.
This does not mean that children of Neanderthals and modern humans must have never been raised by Neanderthals. Swiss population geneticist Laurent Excofier has always been interested in the data from our research group. In 2008, he published a paper on gene flow in which both parties who had gene flow had expanded, while the other party did not expand or even shrank. In this case, genetic variations exchanged between populations are more likely to be preserved in expanding populations than in shrinking populations. If gene exchange occurs at the "outpost" of population development, and this population is showing an expansion, then the exchange-derived variants may occur very frequently in the population. Excofier once vividly and appropriately called this phenomenon "allelic surfing": alleles enter a surging wave of population expansion, causing their frequency to surge. This means that hybridization may have happened in both directions, but we have not found this phenomenon in Neanderthals, because the latter's population size has been reduced after the encounter between modern humans and Neanderthals.
Another more realistic reason why we cannot detect gene flows from modern humans to Neanderthals: Neanderthals in Vandia’s cave lived 38,000 years ago, when hybridization had not occurred. Maybe we will never know the details of Neanderthal mating with modern humans, but that doesn't bother me. For me, the "who has sex with whom" in the late Pleistocene is a secondary issue. Importantly, Neanderthals really have genetic contributions to today's human beings, which is a question about the genetic origin of today's human beings.
After confirming the results of David and Nick's research, we began to explore: how many people outside Africa have genomes from Neanderthals. This result cannot be directly estimated by the SNP match ratio, because the number of matches between Neanderthals and people outside Africa depends on many other variable factors: one is the age in which the common ancestors of Neanderthals and modern humans lived, the other is the time when they crossed, and the third is the size of the group that Neanderthals used to be. Monty Slatkin estimates the proportion of Neanderthal DNA among today’s humans by simulating the population history of Neanderthals and modern humans. The results of its research show that Europeans or Asians inherit about 1% to 4% of Neanderthal DNA. David and Nick did different analyses. The essential question they explored was: how far is Europeans and Asians from 100% Neanderthals - the answer is 1.3%~2.7%. Therefore, we concluded that less than 5% of the DNA of people outside Africa comes from Neanderthals. Although this ratio is small, it is clearly visible.
The last question of this round of work: In addition to inheriting it to Europeans, how did the Neanderthals pass on DNA to the Chinese and Papuans. As far as we know, the Neanderthals have never been to China and must have never been to Papua New Guinea, so we infer that the Neanderthals must have met the ancestors of the Chinese and Papuans somewhere more westward.
We sat around the speaker phone in my Leipzig office during the weekly conference call.I will first suppress the Middle East’s ideas so that the members of the alliance can explore all possibilities with a keen mind. Monty came up with a complex scenario to explain the mutated pattern we see. First, he believed that the Neanderthal ancestors originally lived in a corner of Africa, and later left Africa, and about 400,000 to 300,000 years ago, they evolved into Neanderthals in the western part of Eurasian . Then, 200,000 years later or later, where Neanderthal ancestors originated in Africa, modern human ancestors emerged. If the various groups of Africa remained separate during this period, then the difference in allelic frequency will continue from the departure of Neanderthal ancestors until the ancestors of modern humans begin to spread and expand; then, after the emergence of modern humans, they not only walked out of Africa, but also traveled through Africa, and crossed with ancient Africans, obtaining mutations from early humans. Then, as we can see, Neanderthals are more similar to those outside Africa than Africans.
Although this situation is theoretically possible, the premise is that African populations have remained stable and separated for tens of thousands of years. As Monty himself pointed out, this seems impossible, because humans can easily move around. The more serious problem is the complexity of the past. To rebuild the past, it is best to consider the simplest pattern, even many other more complex situations are likely to happen. Adopt the principle of the simplest explanation of preference, the so-called Parsimony principle. For example, one assumes that the ancestors of Neanderthals and modern humans originated in Asia, and then the ancestors of modern humans went to Africa, leaving no descendants in Eurasia, and then expanded again and replaced Neanderthals. This hypothesis does match all observations, but this hypothesis involves many population flows and population extinction. Relatively speaking, the hypothesis that Neanderthals originated in Africa is much simpler. Therefore, compared with the African origin hypothesis, the Asian origin scenario is not that simple, so it is not the best explanation at present. Even though the African substructure scenario provides possible explanations for our data, it still doesn't hold because there is a simpler and more obvious explanation. In fact, this explanation is too obvious. Several of us thought at the same time: That is the Middle East scenario.
Source: (Sweden) by Svanter Parbo. Neanderthal [M]. Hangzhou: Zhejiang Education Press, 2018.12.
