Contesting the presence of wheat in the British Isles 8,000 years ago by assessing ancient DNA authenticity from low-coverage data.

Clemens L Weiß,Kay Prüfer,Hernán A Burbano,Michael Dannemann

doi:10.7554/elife.10005

Clemens L Weiß, Kay Prüfer + Show 2 more

Open Access

https://doi.org/10.7554/elife.10005

Copy DOI

Abstract

Contamination with exogenous DNA is a constant hazard to ancient DNA studies, since their validity greatly depend on the ancient origin of the retrieved sequences. Since contamination occurs sporadically, it is fundamental to show positive evidence for the authenticity of ancient DNA sequences even when preventive measures to avoid contamination are implemented. Recently the presence of wheat in the United Kingdom 8000 years before the present has been reported based on an analysis of sedimentary ancient DNA (Smith et al. 2015). Smith et al. did not present any positive evidence for the authenticity of their results due to the small number of sequencing reads that were confidently assigned to wheat. We developed a computational method that compares postmortem damage patterns of a test dataset with bona fide ancient and modern DNA. We applied this test to the putative wheat DNA and find that these reads are most likely not of ancient origin.

Highlights

The evolutionary reconstruction of the past has been greatly enriched by direct interrogation of ancient DNA from plants and animal remains (Shapiro and Hofreiter, 2014)
If contamination is a possible explanation for the result, it is crucial to exclude this possibility by giving positive evidence for the authenticity of aDNA (Prufer and Meyer, 2015)
The excess of cytosine to thymine (C-to-T) substitutions at the 5′ end occurs at different magnitudes in samples of different ages, the exponential increase of substitutions towards the end is a ubiquitous pattern in aDNA studies (Sawyer et al, 2012)

Summary

Introduction

The evolutionary reconstruction of the past has been greatly enriched by direct interrogation of ancient DNA (aDNA) from plants and animal remains (Shapiro and Hofreiter, 2014). We present an approach that compares the pattern of C-to-T substitutions in a set of test reads with the distributions of C-to-T substitutions in reads from known ancient- and modern-DNA and apply this approach to sedaDNA from Smith et al Results and discussion

Results

Conclusion