A genomic dating tool for ancient genomes resolves the origins of hundreds of Eurasian genomes

Abstract

Radiocarbon dating is the gold-standard in archaeology to estimate the age of skeletons, a key to studying their origins. Nearly half of all published ancient human genomes lack reliable and direct dates, which results in obscure and contradictory reports. Here, we developed the Temporal Population Structure (TPS), the first DNA-based dating method for ancient genomes ranging from the Upper Palaeolithic to modern-day samples and applied it to 1559 ancient Europeans, Asians, and Caucasus individuals and to 2117 modern worldwide individuals. We show that TPS predictions for dated skeletons align with their known dates and correctly account for kin relationships. The TPS-dating of poorly dated Eurasian samples resolves conflicts and sheds new light on disputed findings, as illustrated by four test cases. We discuss the phenotypic traits of the Time Informative Markers (TIMs) that underlie TPS. SummaryTPS is a novel method to date humans from the Upper Palaeolithic to modern time from their DNA sequences. Accurate dating is essential to the interepretation of paleogemonic data.. The gold-standard method in archaeology is radiocarbon dating1. However, a major limitation of radiocarbon dating is the high amount of collagen extraction (500 mg) involved in the process2. Consequently, half of all published ancient human genomes lack reliable and direct dates, which results in obscure and contradictory reports. Here, we present the Temporal Population Structure (TPS), the first genomic dating method for ancient genomes ranging from the Upper Palaeolithic to modern-day samples. We show that TPS predictions for 961 radiocarbon-dated Eurasian skeletons align with their known dates. We replicate these findings on 598 other Europeans, Asians and Caucasus individuals. Using kin-pairs, we demonstrate that TPS has produced more accurate results than radiocarbon and other dating. We show how our findings resolve conflicts and sheds new light on disputed findings as illustrated by four test cases. Finally, we discuss the phenotypic traits of the Time Informative Markers (TIMs) that underlie TPS. TPS is a novel dating technique, which can be used when radiocarbon dating is unfeasible or uncertain or to develop alternative hypotheses. TPS cannot be used for older (<14,000 years ago) samples, and its accuracy depends on the temporal and geographical breadth of radiocarbon-dated samples in the training dataset, though this limitation can be improved over time. Overall, TPS can improve the accuracy of archeological and paleogenomic studies.