930000 years ago, there were only over a thousand human ancestors left? Humanity has always been a community of shared destiny!

Author: Zhang Jiansong, Xinhua Daily Telegraph reporter

The human genome is an autobiography written using genetic codes. Since the birth of life, genetic language has been used to record the changes and vicissitudes of human life.

Chinese scientists have innovatively constructed whole genome analysis methods and theories, and found that 930000 years ago, during the transition period between the early and middle Pleistocene, human ancestors lost 98.7% of their member individuals in a short period of time due to severe climate change; In the following 117000 years, the average number of adult individuals in human ancestors was only 1280. The international academic journal Science published relevant research papers online on September 1st.

930000 years ago, humans encountered a serious "group bottleneck" and were almost extinct? This research achievement has attracted widespread attention and even questioning from various sectors of society. Our reporter interviewed Li Haipeng, the co corresponding author of the paper and researcher of the Chinese Academy of Sciences Shanghai Institute of Nutrition and Health.

Research Results Diagram

The "group bottleneck" coincides with a series of evidence from anthropology and paleoclimatology over time

"Who am I? Where do I come from? Where am I going?" Faced with these three ultimate questions, scientists have never stopped exploring the mystery of the origin of humanity itself.

According to current archaeological findings, human evolution began with the forest apes of primates, and then went through four stages: southern apes, Homo sapiens, Homo erectus, and Homo sapiens. Among them, Homo sapiens is the surviving human species in the world. The past one million years have been a critical period of human evolution, but due to limitations in analytical methods, research on the history of changes in the number of human ancestor populations is mostly limited to the last 300000 to 100000 years.

Fossils are the most direct evidence for studying human evolution. Paleontologists have discovered that there are "missing links" in the fossils of African ancient humans. Fossils between 950000 and 650000 years ago are extremely scarce, but those before 950000 and within 650000 years ago are relatively abundant. Multiple fossils of Homo erectus discovered in Africa are all about 900000 years ago, while those within 900000 years ago have not been discovered. Instead, a new human species has emerged.

"We have analyzed the modern human genome and found that human ancestors experienced a population bottleneck of 117000 years between 93000 and 813000 years ago. This period of time perfectly explains the missing link in African human ancestor fossils and the mystery of the disappearance of African Homo erectus fossils. Because there are no fossils left," said Li Haipeng.

According to research results, approximately 813000 years ago, human ancestors quickly recovered from the process of population decline, corresponding to the formation of a new ancient human species.

"Based on our analysis, this is a new ancient human species, rather than one ancient human species replacing another." Li Haipeng said, "How to name this new ancient human species? There is currently a lot of controversy in the fossil world. To avoid controversy, we have named it LCA, the closest common ancestor of modern humans, Neanderthals, and Denisovans."

The change in chromosome number has a significant impact on organisms. The fusion or cleavage of chromosomes is an important event in the evolutionary process. Human beings are descendants of apes and share a common ancestor with chimpanzees. But since chimpanzees and humans parted ways on the evolutionary tree 6 million years ago, there has been a chromosomal fusion event in the human branch that formed the modern human chromosome 2, which is a fusion of two medium-sized ape chromosomes.

"By calculating the conversion rate from AT to GC in the fusion region of human chromosome 2 and comparing it with the homologous region of chimpanzees, the fusion time of two ancient human chromosome 2 can be estimated. Two research groups have estimated the fusion time of ancient human chromosome 2. One group of studies estimated it to be 740000 years ago, while the other group estimated it to be 900000 years ago. Therefore, the fusion time of this chromosome is likely to occur between 740000 and 900000 years ago, which corresponds to the severe bottleneck period we studied in ancient times." Li Haipeng said.

Distribution map of African family fossils

Innovatively constructing the theory and method of "fast and minimal time tracing"

In recent years, although ancient DNA sequencing technology has developed rapidly, scientists are unable to extract ancient DNA from African human ancestor fossils 300000 years ago due to hot conditions that are not conducive to DNA preservation. The achievement of this research is attributed to Li Haipeng and Pan Yixuan from the Institute of Brain Functional Genomics at East China Normal University, leading team members such as Hu Wangjie and Hao Ziqian to innovate and construct a new theory of "rapid and minimal time tracing".

This new theory, based on the intersection of population genetics and computational biology, can achieve high-precision time tracing on a large time scale of millions of years, accurately estimate the history of ancient populations, and thus conduct a population census of ancient human populations.

Although there is no written record of the number of populations in prehistoric times, the size of the effective population can affect the "tracing rate" of each generation, which is the probability that two lineages come from the same ancestor in the previous generation. Therefore, human ancestors left imprints in the population genome, reflecting the size of the population at that time. The older the group's history, the weaker the imprint signal that remains to this day.

Li Haipeng and Pan Yixuan's research team created the FitCoal theory based on this principle. Following this theory for mathematical derivation, obtaining analytical solutions for the long-term expected values of ancestral branches corresponding to each mutation type in any population model can yield accurate likelihood values, that is, the probability of observing the sample mutation spectrum under the historical conditions of the population. Therefore, without prior knowledge of group history, FitCoal can automatically and quickly search for maximum likelihood values, thereby estimating group history.

The research encountered significant difficulties at the beginning. "Our approach is to infer population history by calculating the expected value of the mutation spectrum under any model and comparing it with the observed mutation spectrum. The expected value of the mutation spectrum is equal to the product of the expected value of the branch length corresponding to the ancestral tree and the mutation rate. However, we initially had no idea how to calculate the expected value of the branch length corresponding to the ancestral tree," said Li Haipeng.

After nearly a year of arduous mathematical deduction and computer programming, researchers have found themselves trapped in an endless quagmire, unable to see where the way out lies. The direction of theoretical deduction was found to be incorrect, the program memory cost was extremely high, and it had no practical value at all. Moreover, the calculation results were also highly erroneous. After many discussions, Pan Yixuan suggested, "Instead of spending time unraveling a tangled knot, it's better to rewind a new knot."

"The research team ultimately decided to delete all the code, start mathematical derivation from scratch, and start programming from scratch. Afterwards, the research returned to the right track without encountering too many major problems, and through daily discussion mechanisms, many detours were avoided. The process of scientific research is a combination of pain and joy," said Li Haipeng.

On August 31, the Shanghai Institute of Nutrition and Health of the Chinese Academy of Sciences held a press conference on major scientific research achievements. Photo by Zhang Jiansong, Xinhua Daily Telegraph reporter

Verify the research results for seven years and publish a preprint of the paper

After three to four years of constructing a new theoretical method of "fast and minimal time tracing", the research team began conducting population data analysis from 2016 to 2017 and quickly "saw" the bottleneck of ancient human populations.

Over a period of 117000 years, the average number of adult individuals in human ancestors was only 1280? "Looking back, when we first saw this result, we were not surprised, but a bit disappointed. Because it was so unbelievable, perhaps it meant that our program was wrong. After all, this is an obvious group historical event. Why did our predecessors not notice it? Could it be that we miscalculated somewhere?" said Li Haipeng.

Team members such as Hu Wangjie, Hao Ziqian, Li Haipeng, and Pan Yixuan repeatedly checked theoretical methods and code. After several months and countless attempts, they still couldn't find any errors. At this point, they realized that the bottleneck of this human community may actually exist.

Afterwards, they spent nearly seven years confirming the existence of this group bottleneck. From analyzing genomic data from thousands of people to analyzing data from another human genome diversity project with a high sequencing rate but relatively small sample size, the same population bottleneck was detected. And the estimated bottleneck time and number of individuals obtained from two independent sets of genomic data from 50 modern human populations are almost identical.

Subsequently, the researchers further validated using two southern African populations from the HGDP-CEPH dataset. Although the sample size was only 6 and 8 individuals, the bottleneck of the ancient population was still detected. The research results of resampling samples from African populations indicate that this ancient population bottleneck can be detected with only the genomes of three African individuals, further indicating the significant discovery brought about by the innovation of the "fast and minimal time tracing" theory.

"In the field of population history estimation, population genetics has a standard falsification process. It can be simply understood that the change in population size is a statistically significant event that occurs, equivalent to a hypothesis test. Therefore, we have verified our conclusions through a large number of simulation experiments. The appendix of the paper has more than 100 pages, and a large part of it is the verification of population bottlenecks in ancient times, all of which support our conclusions." Li Haipeng said.

He said, "Over the past decade, the vast majority of our time has been focused on our discoveries and further verification. Just like Holmes' famous saying: eliminate everything that is impossible, and the rest, even if it is unbelievable, is the truth. The criticisms and suggestions currently appearing on the internet have not exceeded the scope we have considered in the repeated verification process over the past six or seven years."

Despite repeated cross validation confirming the discovery of an important human population bottleneck, there is still a lack of paleontological evidence. After discussion, the research team decided in May 2021 to openly listen to feedback from the academic community and seek cooperation by releasing a preprint version.

A few days after the preprint was released, paleoanthropologists Giorgio Manzi and Fabio Di Vincenzo from the University of Rome and the University of Florence in Italy quickly contacted and informed them that this discovery was consistent with their fossil archaeological findings. After careful discussion and exchange, everyone has decided to unite and further complete this project. Finally, after nearly a year of close collaboration and revision, the article was sent to Science on April 28, 2022.

In the officially published paper, the research team included numerous technical details in over 100 pages of supplementary materials. In order to facilitate the academic community to replicate and verify their results, specific parameters of the simulation program have been specially provided and formatted for peer use.

Professor and doctoral supervisor Wu Zhongyi from the School of Life Sciences at Sun Yat sen University commented, "The theoretical framework of this paper is all in the attachment. The main text only publishes the conclusion. This is a story of buying a box and storing pearls, and it is also one of the rare theoretical papers in CNS journals in recent years."

On August 31st, Researcher Li Haipeng presented his research findings at a press conference. Photo by Zhang Jiansong, Xinhua Daily Telegraph reporter

Humanity has always been a community of shared destiny since ancient times

The evolutionary information traced back through genomic data of modern populations represents the information of modern human ancestors.

"The bottleneck observed in our study in ancient populations represents the direct ancestor of modern humans, which quickly lost 98.7% of its member individuals in a very short period of time, decreasing to 1.3% of its original population. This is a very strong impact, consistent with the observed population changes in many endangered species. The conclusion that human ancestors were once on the brink of extinction does not rely on the molecular clock hypothesis," said Li Haipeng.

He believed that at that time, if human ancestors had not taken any practical action, made no changes, and allowed the population to rapidly decline, it would have been the result of the extinction of the entire population. Only by closely uniting and obtaining and distributing extremely limited food more effectively and fairly can our ancestors overcome difficulties over a period of 117000 years.

In addition to cultural and collaborative changes, survival pressures have also led to many evolutionary improvements in human ancestors. For example, an increase in brain capacity, fusion of two chromosomes, and the formation of new ancient human species. In addition, the early archaeological records of human fire control are very close to the time when the human ancestor population began to recover rapidly, and may also play an important role in restoring population size.

"Since ancient times, humanity has been a community with a shared future. Faced with the harsh living environment, our ancestors united and responded, which led to the prosperity of human life on Earth today." Li Haipeng said, "Today, facing severe extreme weather and complex disputes, we should return to the origin of humanity, uphold the concept of a community with a shared future for mankind, unite and assist, and develop together, in order to better reproduce and thrive for generations on Earth."

Professor Liu Xiaoming, a human geneticist at the University of South Florida, Professor Yang Jian, the head of the Statistical Genetics Laboratory at West Lake University, and Professor Yang Jian from the School of Life Sciences, among other experts, believe that the milestone event of "population bottleneck" that humans encountered 930000 years ago has a significant impact on human evolution and may have determined the formation of many key phenotypes in modern humans. The sharp decline in the population size of human ancestors during this ancient period has also reduced the genetic diversity of modern populations by 65.85%, and has had a profound impact on human life and health. This is a research paradigm where new theories bring new discoveries.