Computational Ancient Genomics (Janet Kelso)
The group develops and applies computational approaches for the analysis of ancient and modern human genomes.
In the last 10 years ongoing work in the Genetics department has made it possible to retrieve and sequence DNA from ancient biological samples dating to as much as 400,000 years old. We have sequenced whole and partial genomes of a number of Neandertals, as well as from early modern humans, and continue to work towards adding new archaic and early modern human genomes in order to improve our understanding of the relationships, population histories and interactions of early human groups.
A number of challenges accompany the analysis of such ancient samples. Only very small amounts of endogenous DNA are typically present in such old materials, and the vast majority of the DNA obtained is from micro-organisms. In addition, the remaining DNA is usually degraded into very short, chemically damaged fragments. We therefore apply a combination of molecular and computational approaches to optimize the retrieval of DNA and to ensure that the sequences that are used in analysis are reliable.
Efforts in the group focus on developing software and analysis methods for the analysis of ancient DNA sequences. These methods are constantly being updated and improved in order to improve our ability to work on even older and more poorly preserved specimens. We make all our software available under open source licenses.
One of the most striking findings of the Neandertal genome project was the discovery that Neandertals contributed approximately 2% of the genomes of present-day non-Africans, and that Denisovans contributed approximately 4-6% of the genomes of people who today live in Oceania. We have had a particular interest in understanding the timing and effects of interactions between Neandertals, Denisovans and early modern humans. Our most recent work has used a simple computational approach to identify haplotypes that are likely of archaic origin and to combine these with large public human genome, gene expression, and phenotype datasets to determine the likely functional consequences of these haplotypes. Our work has shown that Neandertal DNA influences a number of traits such as immunity and those influenced by sun-exposure.
Software and Databases
We develop software and databases for the analysis of ancient and modern genomes. A list of published software packages is available here:
Or on our group’s Github site https://github.com/mpieva
All our software are available under open source licenses.