Periodic Reporting for period 1 - NEADMIX (Inferring hominin population history through space and time using introgressed haplotypes)
Reporting period: 2022-05-01 to 2024-10-31
Our closest evolutionary relatives, Neandertals and Denisovans, died out roughly 40,000 years ago. Understanding how we differ from them is a major goal of paleoanthropology. Technical improvements in the retrieval of DNA from fossil bones, teeth and sediments have recently provided access to genetic data from the Middle and Upper Palaeolithic, the crucial time period where early modern humans expanded out of Africa and met Neandertals and Denisovans. The genetic study of these time periods is mainly limited by sample availability: under ideal conditions, DNA is preserved for up to two million years, but preservation varies highly between regions and climatic conditions; in Europe, DNA older than 40,000 years old is rare. Thus, despite advances in sequencing technologies and ancient DNA protocols, many samples that come from the time when both Neandertals and modern humans lived in the same place have little DNA.
The key objective of the project is to learn on when, how, and where Neandertal, Denisovans and modern humans met. We want to learn about their interactions, and what the biological consequences were of the gene flow. To do this, we need to analyze genetic data from people that lived (nearly) contemporaneous to Neandertals and Denisovans. This requires the development of novel theory, and their implementation in new computational and statistical method to analyze ancient DNA. To achieve this, we will develop novel population genetic theory, first in the simplified case of a single, neutral population, but later extend this to incorporate both natural selection and population structure. This will be then applied in a software that aims at characterizing the parts of an individuals genome they have inherited from a Neandertal and Denisovan ancestor. Our tools will be targeted towards addressing the issues with ancient DNA, namely that they are often low quality, that the DNA is highly fragmented and damaged, and very frequently contaminated.
We will then apply our tools to a comprehensive data set of present-day and early modern human genomes. We will generate comprehensive genetic maps, flagging for each individual where in its genome it has genetic material inherited from Neandertals or Denisovans. We will then use these data to precisely date when, and over which duration, humans met with Neandertals and Denisovans. We will also be able to say which regions were under positive or negative natural selection, and over which time periods this took place. We will also investigate how these patterns change over time and space. Overall, this study will give us unique and novel insights into our relationship with Neandertals and Denisovans, and so clarifies what makes us human.
The key objective of the project is to learn on when, how, and where Neandertal, Denisovans and modern humans met. We want to learn about their interactions, and what the biological consequences were of the gene flow. To do this, we need to analyze genetic data from people that lived (nearly) contemporaneous to Neandertals and Denisovans. This requires the development of novel theory, and their implementation in new computational and statistical method to analyze ancient DNA. To achieve this, we will develop novel population genetic theory, first in the simplified case of a single, neutral population, but later extend this to incorporate both natural selection and population structure. This will be then applied in a software that aims at characterizing the parts of an individuals genome they have inherited from a Neandertal and Denisovan ancestor. Our tools will be targeted towards addressing the issues with ancient DNA, namely that they are often low quality, that the DNA is highly fragmented and damaged, and very frequently contaminated.
We will then apply our tools to a comprehensive data set of present-day and early modern human genomes. We will generate comprehensive genetic maps, flagging for each individual where in its genome it has genetic material inherited from Neandertals or Denisovans. We will then use these data to precisely date when, and over which duration, humans met with Neandertals and Denisovans. We will also be able to say which regions were under positive or negative natural selection, and over which time periods this took place. We will also investigate how these patterns change over time and space. Overall, this study will give us unique and novel insights into our relationship with Neandertals and Denisovans, and so clarifies what makes us human.
In the first two years of the project, we have developed and implemented a novel algorithm to map Neandertal and Denisovan ancestry in early modern humans, in a program called admixfrog. While still under development, admixfrog is currently heavily used by the team members, as well as some outside groups. We also have made good progress on studying the population structure and population history of Neandertal ancestry in early Eurasians. In a recent preprint, we find that most Neandertal ancestry is shared between all modern humans outside Africa, and that it traces back to a single gene flow event. We find that the peak of the gene flow was around 47,000 years ago, and that it took place over several thousand years. We also find that most natural selection-both positive and negative- happened immediately after the gene flow. In addition, we are also developing new methods to detect Denisovan ancestry. Since Denisovan ancestry is rarer, and difficult to detect, we are developing and evaluating a number of new algorithms that do not use an archaic reference genome. We find that our approach works well on genomes at coverages around 3X, and are currently testing and implementing it on simulations.
Questions about our origins, our prehistory, how humans came to be have been asked for millenia. This is reflected by a very high public interest in research in that field. Particularly genetics, using ancient DNA, is a new tool that can provide novel and different insights compared to other archaeological techniques. As such, our work - both related to this project, and other - have been frequently covered by news outlets. Indeed, even our preprint from this project has received attention from multiple news outlets - both from science media and mainstream media, showing the importance and extremely high interest in this type of work. Overall, the computational and statistical methods that we are developing are pushing the boundaries on what we can do with ancient DNA, increasing the reach and utility of this approach. This is particularly important because DNA degrades much faster in hotter and more humid climates - regions that are often less involved in scientific research in the first place.