AI explores ‘dark genome’ to shed light on cancer growth

Scientists have used artificial intelligence to illuminate the human body’s so-called dark genome and develop a potentially powerful new approach to cancer detection, monitoring and treatment.  

Investigators at Johns Hopkins University in the US have revealed how previously little-studied repeats of genetic code sequences — known as “junk DNA” or dark matter — are associated with tumours.

The research, published in Science Translational Medicine on Wednesday, shows how AI and other advanced computing technologies are deepening understanding of diseases and how to deal with them.    

“It’s like a grand unveiling of what’s behind the curtain,” said Victor Velculescu, co-author of the paper and an oncology professor at the Johns Hopkins Kimmel cancer centre, adding that it will “lead to new therapies, new diagnostics and new screening approaches for cancer”.

The researchers explored the “dark matter” of genetic code that — like its astronomical namesake — is detectable but mysterious and beyond the scope of previous conventional analysis.

They used a machine learning technique known as Artemis, named for the Greek hunting goddess, to analyse junk DNA found in tumours or floating in the bloodstream after being shed from cancerous growths. 

The scientists reviewed more than 1,200 distinct types of repeat DNA elements comprising almost half the human genome. They found that many of these junk genetic code pieces not previously linked to cancer had changed during tumour formation. 

In the researchers’ analysis of 736 genes linked to cancer formation, almost two-thirds contained on average 15 times more the number of repeat DNA sequences than expected. The findings suggested the junk pieces were “not occurring randomly” in cancerous tissues and were instead potentially revealing markers of malignance, Velculescu said.

“They end up being clustered around genes that are altered in cancer in a variety of different ways, providing the first glimpse that these sequences may be key to tumour development,” he said.  

The research is the latest effort to use cutting-edge computing techniques to boost the fight against cancer, which causes immense suffering worldwide and is expensive for health services to treat. Pharmaceuticals companies are using AI to identify new molecules as targets for possible anticancer drugs.

Other scientists have deployed the technology in an effort to sharply cut the number of missed early-stage breast cancers. 

The charity Cancer Research UK welcomed the Johns Hopkins paper and the further evidence it offered about the promise of technologically assisted tumour detection. 

“While still at an early stage, this research demonstrates how some cancers could be diagnosed earlier by detecting tumour-specific changes in cells collected from blood samples,” said Hattie Brooks, the charity’s research information manager.

“Although there is still a long way to go before Artemis could be used in the clinic, it’s always exciting to see new approaches showing potential in detecting cancer sooner.”