Earlier this year, a publication in the prestigious journal Science claimed evidence for the trade of wheat between English hunter-gatherers and more advanced European farmers about 8,000 years ago – 2,000 years before Britons were practicing agriculture. This meant that at that time, Mesolithic Britons may have been importing their cereals from Neolithic communities before cultivating them themselves.
Now, eight months on, scientists are contesting it, suggesting that the samples may not be from ancient times, but are instead most probably modern contaminants.
“I cannot say these samples are definitely not of ancient origin,” lead author Hernán Burbano from the Max Planck Institute of Developmental Biology told IFLScience. “That’s unscientific. I’ve done a statistical test and shown that they lie outside the distribution of ancient DNA and are therefore most likely not ancient DNA.”
The controversy centers around something known as cytosine deamination, which is where one of the building blocks of DNA – cytosine – gets substituted for a different base, called uracil. These biochemical changes tend to affect the ends of DNA fragments and accumulate with age, so the researchers behind the current study, published in eLife, wanted to compare substitution patterns from ancient and modern samples as a method of determining the authenticity of ancient DNA.
The team looked at a plethora of randomly selected plant and animal DNA from different ages, and also compared the wheat used in the Science study, gathered from submerged sediment off the Isle of Wight, with relatively modern samples collected from potato leaves.
The problem with ancient samples is that fragile DNA fragments over time. Water, oxygen, sunlight, and microbes can all speed up this process, making old DNA difficult to read by sequencing techniques. This left the original team, from the University of Warwick, with just 150 sequences to work with, so Burbano’s group selected samples that also had the same number of reads. Using this data, Burbano concluded that the “damage” or substitution patterns of the ancient samples didn’t match up with what would be expected from samples of this age.
“It’s a pretty weak attack,” Robin Allaby, lead researcher of the Science study, told IFLScience. “They haven’t taken into account thermal age,” referring to the fact that breakdown processes are slowed down by lower temperatures. “The wheat samples came from sediment that has been preserved in a constant four degrees Celcius [39.2 degrees Fahrenheit] for almost 8,000 years,” he added. While Allaby admits that scientists don’t have a great understanding of the rate of deamination with temperature, he says we do know that colder samples, specifically those from permafrost, have lower levels of deamination.
“Our best guess for what the conversion rate would be expected to be at around 8,000 years is about 5 percent,” argued Allaby. “If that’s the case, then in 150 sequences we would be looking for about four or five cytosines that have been converted. It’s highly unlikely this would even be detectable.”
But Burbano counters this argument, stating that high levels of damage patterns are clearly present in colder samples, such as Ötzi the Iceman who is more than 5,000 years old. “You don’t need to match the conditions on the sample,” he said.
Evidently the argument is not over, and Allaby’s group is now preparing a published response. But at the very least, “it opens up a wider debate that hasn’t really been addressed properly, this broader understanding of cytosine deamination,” he added.