Back to overview

Better genotype likelihoods for ancient DNA

English title Better genotype likelihoods for ancient DNA
Applicant Wegmann Daniel
Number 200420
Funding scheme Project funding (Div. I-III)
Research institution Département de Biologie Faculté des Sciences Université de Fribourg
Institution of higher education University of Fribourg - FR
Main discipline Genetics
Start/End 01.06.2021 - 31.05.2025
Approved amount 761'798.00
Show all

All Disciplines (2)


Keywords (5)

Statistical Genomics; Bioinformatics; Genotype Likelihoods; Migration Period; Ancient DNA

Lay Summary (German)

Aus Fossilien gewonnene DNA gibt uns Einblicke in vergangene Gesellschaften und Kulturen. Solche alte DNA ist aber stark beschädigt und kann meist nur in Kleinstmengen gelesen werden. Dieses Projekt entwickelt statistische Verfahren um mit diesen Problem umzugehen.
Lay summary
Inhalt und Ziele des Forschungsprojekts

Alte DNA ist wie eine Zeitmaschine welche es erlaubt, genetische Proben aus vergangener Zeit zu analysieren und mit Proben aus anderen Epochen zu vergleichen. Solche Vergleiche geben Aufschluss über die evolutionäre Geschichte einer Art oder Population, uns ebenso über unser eigene Geschichte. Alte DNA ist aber chemisch beschädigt und lässt sich auch mit modernsten Technologien oft nur in Kleinstmengen  gewinnen. Die resultierende Unsicherheit lässt alte Proben deutlich anders erscheinen als moderne und muss daher statistisch herausgerechnet werden. Hierfür entwickelt dieses Projekt neue statistische Verfahren und entsprechende Computerprogramme. Diese Verfahren werden dann genutzt um zu untersuchen, inwiefern die Verbreitung der Slawischen Kultur auf eine Wanderbewegung zurückzuführen ist.

Wissenschaftlicher und gesellschaftlicher Kontext des Forschungsprojekts

Dieses Projekt befasst sich mit Grundlagenforschung. Die entwickelten Verfahren werden aber eine breite Anwendung finden und die Forschung mit alter DNA beschleunigen. Diese Forschung revolutioniert gerade unser Wissen über unsere eigene Geschichte von der Frühzeit bis heute, und damit auch unser Selbstverständnis.

Direct link to Lay Summary Last update: 07.04.2021

Responsible applicant and co-applicants


Project partner

Natural persons

Name Institute

Associated projects

Number Title Start Funding scheme
173062 Identifying selection in the presence of gene flow 01.06.2017 Project funding (Div. I-III)


Recent technological and methodological improvements now make it possible to routinely extract and sequence DNA of ancient and historical samples, and hence to provide a complementary angle on many long-standing questions both in evolutionary biology as well as in history and archaeology. Comparing population samples obtained from different time points, for instance, allows to track and study evolutionary processes even in long-lived species. In the current situation of a global pandemic, it seems particularly easy to appreciate the importance of aDNA in understanding the etiology of infectious diseases and the evolution of their pathogenicity. Ancient DNA (aDNA) has also significantly contributed to our understanding of many important periods of the human past and our understanding of human evolution itself.The bioinformatic analysis of aDNA, however, remains challenging due to the decay of DNA over time. As a result, obtained aDNA sequences are usually short, affected by post-mortem-damage (PMD) and sequenced alongside a large fraction of contaminating sequences. While rigorous lab practices help reduce these effects, any inference of genetic relationship and diversity among past samples must account for them, as well as sequencing errors.Traditionally, this has been done by strict filtering, discarding potentially damaged bases, and restricting the analysis to a subset of the genome. Here I argue that the power of aDNA inferences can be much improved by switching to methods that account for these processes probabilistically, rather than by discarding data, and I propose to develop a flexible and modular framework to do so.This is now possible thanks to many important statistical developments, namely the introduction of genotype likelihoods to summarize genotyping uncertainty, the development of many tools to infer population genetic quantities from genotype likelihoods, and the development of many statistical models to quantify contamination and sequencing error rates from aDNA sequence data. I propose here a general, modular statistical framework to calculate genotype likelihoods for aDNA that builds on these individual models. To allow for fast and reproducible processing of many samples, I further propose to embed this statistical framework into an automated pipeline, similar to those recently developed for existing methods. The proposed statistical framework builds on several recent developments in my group: a genotype likelihood model accounting for PMD, methods to infer PMD and sequencing error rates and the development of a bioinformatic toolbox to work with aDNA. As I argue here, this toolbox can be generalized to incorporate additional models accounting for all processes affecting aDNA data, thereby greatly improving the power of downstream analyses.I further propose to make use of these tools to characterize the importance of human migration during the genesis of the early Slavic culture. While Slavic languages are widespread in Europe today, surprisingly little is known about the early history of Slavic people. In particular, it remains unclear whether the expansion of the Slavic culture since the 5ths century involved the expansion and migration of people - or whether it was rather the result of a pure cultural diffusion. I propose here to shed some light on this questions by focusing on Breclav, a small geographic region in Czechia with particularly well preserved inhumations of both pre-Slavic and Slavic cultures that are exceptionally well characterized in terms of their archeological context, also in terms of social structures. By focusing on a small region, we boil down the question of a population turnover to a well defined, archaeologically contextualized, geographically and chronologically framed question that I will address using explicit hypothesis testing: what was the level of genetic continuity between nearby archeological sites? Precisely because of its well defined localized focus, this study will contribute substantially to our understanding of the genesis of the Slavs, and as such contribute to the timely discussions on the identity of Europeans and European nations.