Scientists have used human cells fused with plant DNA to test the purpose of non-coding DNA in the human genome. The experiment shows that activity in this so-called 'dark DNA' is largely random noise, supporting the idea that much of it is junk. The findings challenge claims that high activity implies function.
For decades, biologists have debated the role of non-coding DNA, which makes up about 98.8 percent of the human genome. While a 2012 project called ENCODE suggested over 80 percent of the genome is active—implying importance—others argued this activity could be mere noise.
In 2013, Sean Eddy at Harvard University proposed inserting random synthetic DNA into human cells to establish a baseline for activity. Making large amounts of synthetic DNA proved costly, but researchers Brett Adey and Austen Ganley at the University of Auckland in New Zealand found a natural alternative. They analyzed human-plant hybrid cells created in Japan, which contain 35 million base pairs of DNA from thale cress (Arabidopsis thaliana).
Plants and animals diverged from a common ancestor at least 1.6 billion years ago, making the plant DNA effectively random from a human perspective, as mutations have altered every site multiple times. Adey and Ganley measured transcription start sites—points where DNA is copied into RNA—in the non-coding regions. They found about 80 percent as many start sites per kilobase in the plant DNA compared to human non-coding DNA.
"A large amount can simply be explained by background noise," Adey said. "This seems to be broadly consistent with the junk DNA idea."
The results indicate that most observed activity, as reported by ENCODE, is noise without function. "This is an excellent demonstration of how biology is, indeed, noisy," commented Chris Ponting at the University of Edinburgh. Dan Graur at the University of Houston called it "yet more experimental evidence confirming what has been obvious for years: most of the human genome is junk."
Human DNA showed 25 percent more activity, which the team cannot yet explain but plans to investigate using machine learning. The findings have not been published in a paper yet, but they bolster arguments that only about 5 percent of the genome is evolutionarily conserved and functional. Evolution tolerates noise, which can sometimes lead to useful innovations, Ganley noted.
