Evidence for wheat in Stone Age Britain thrown into doubt
A new test reveals that samples of wheat DNA found in 8,000 year old samples are most likely not of ancient origin.
Ancient DNA, that is to say DNA extracted from fossils and ancient remains, provides a window into the past lives of humans, animals and plants. But working with this DNA is challenging; DNA decomposes with time, and so ancient DNA is often fragmented, damaged and present in tiny quantities. Furthermore, ancient DNA is also easily contaminated by modern DNA from those handling it and its surroundings. Researchers have therefore developed special protocols for working with ancient DNA and tests to indicate whether it has been contaminated.
One approach used to check that DNA is of ancient origin identifies a pattern of damage that is specific to ancient DNA. This damage changes the building blocks that make up DNA, causing one (called cytosine or C) to be misread as another (thymine or T). This substitution occurs most frequently at the ends of ancient DNA molecules, and occurs less often along its length, forming a detectable and characteristic pattern of damage.
A common way to analyse ancient DNA is to sequence it and then compare the resulting sequences to the genomes of modern organisms to identify its origins. In a study published earlier in 2015, investigators sequenced the DNA present in sediments obtained from a submerged archaeological site off the coast of the Isle of Wight in the United Kingdom. This previous study identified some DNA fragments that matched sequences in the wheat genome. This led the investigators to conclude that wheat was present in the British Isles around 8,000 years ago, some 2,000 years earlier than previously thought.
However, possibly owing to the small number of fragments that were found, the previous study did not check if the damage pattern matched that expected for ancient DNA. Now, Clemens Weiß and colleagues have developed a new computational method that tests whether DNA shows a typically ancient, or typically modern, pattern of C-to-T substitutions. When this test was used to assess the wheat sequences that were previously claimed to have ancient origins, it revealed that their pattern of DNA damage did not fit statistically with those of ancient DNA.
Weiß and colleagues’ findings contest those of the earlier study, and suggest that the new statistical method could be used to authenticate ancient DNA even when the number of available sequences is low.
To find out more
Read the eLife research paper on which this eLife digest is based: “Contesting the presence of wheat in the British Isles 8,000 years ago by assessing ancient DNA authenticity from low-coverage data” (November 3, 2015).
