Skip to main content

IBM Research is using A.I. algorithms to unlock the secrets of dark matter DNA

You may have heard of dark matter, a mysterious form of lesser-studied matter that is thought to be composed of some as-yet undiscovered subatomic particles, but which makes up an astonishing 85% of the matter in the universe. But how about dark matter DNA? These unexplored molecules and matter surrounding our genes make up more than half of the human genome — but are a total conundrum in terms of what they encode and, more importantly, affect.

The good folks at IBM and the Munich Leukemia Laboratory think they can help come up with some answers — and they’ve used some groundbreaking A.I. algorithms to help.

Recommended Videos

“Despite it making up a large portion of our genome, dark matter DNA has been ignored, as most scientists believe it plays no role,” Laxmi Parida, IBM Research Fellow in Computational Genomics, told Digital Trends. “At IBM Research, we thought there might be more to dark matter DNA than we have been led to believe.”

Please enable Javascript to view this content

The researchers designed what they refer to as a “stochastic regularization A.I. model” that was specifically built for DNA data. Using this model, which they termed ReVeal, the team was able to train algorithms on data from patients’ blood samples; allowing the A.I. to learn from and separate specific signals from the dark matter, as well as the rest of the DNA.

“Our most exciting finding was that using ReVeal, we could achieve a 75% accuracy rate in identifying blood cancers just by looking at the DNA or dark matter DNA alone in a patients’ blood sample,” Parida continued. “This is compared to just a 35% accuracy rate with standard A.I. methods on this data.”

From these findings, the researchers suggest that dark matter DNA plays a much larger role than previously thought in influencing the phenotype of cells and tissues. This suggests that dark matter DNA, far from being an inconsequential research footnote, may have a much greater role to play in our genome than people realized. The results also show that DNA alone contains enough signals to accurately classify blood cancers. This lays the groundwork for further analysis into how blood tests could be used to diagnose these complex diseases — which, up until now, required invasive tissue biopsies, histologies, and more.

“In the long term, we hope this leads to more breakthroughs into how dark matter DNA influences the human genome, and clues it can give us into better understanding our genetics and complex disease,” Parida said.

Luke Dormehl
Former Digital Trends Contributor
I'm a UK-based tech writer covering Cool Tech at Digital Trends. I've also written for Fast Company, Wired, the Guardian…
We used an A.I. design tool to come up with a new logo. Here’s what happened
Digital Trends AI logo

No matter what industry you work in, you’ve probably heard that artificial intelligence is coming for your job. Factory workers, news reporters, even stock brokers have all seen A.I. move into their fields, automating some of their roles. Proponents of automation point out that it tackles the menial, repetitive tasks, freeing workers to focus on more creative aspects.

Now, gig economy marketplace Fiverr recently announced a new A.I.-powered tool that helps businesses create a logo.

Read more
Fake news? A.I. algorithm reveals political bias in the stories you read
newspaper stack

Here in 2020, internet users have ready access to more news media than at any other point in history. But things aren’t perfect. Click-driven ad models, online filter bubbles, and the competition for readers’ attention means that political bias has become more entrenched than ever. In worst-case scenarios, this can tip over into fake news. Other times, it simply means readers receive a slanted version of events, without necessarily realizing that this is the case.

What if artificial intelligence could be used to accurately analyze political bias to help readers better understand the skew of whatever source they are reading? Such a tool could conceivably be used as a spellcheck- or grammar check-type function, only instead of letting you know when a word or sentence isn’t right, it would do the same thing for the neutrality of news media -- whether that be reporting or opinion pieces.

Read more
Barnes & Noble used A.I. to make classic books more diverse. It didn’t go well
Barnes and Noble Diverse Editions

For Black History Month, Barnes & Noble created covers of classic novels with the protagonists re-imagined as people of color. Then it quickly canceled its planned Diverse Editions of 12 books, including Emma, The Secret Garden, and Frankenstein amid criticism that it clumsily altered books by mostly white authors instead of promoting writers of color. The project used artificial intelligence to scan 100 books for descriptions of major characters, and artists created covers depicting Alices, Romeos, and Captain Ahabs of various ethnicities.

"We acknowledge the voices who have expressed concerns about the Diverse Editions project at our Barnes & Noble Fifth Avenue store and have decided to suspend the initiative," Barnes & Noble announced in a statement. The company partnered with with Penguin Random House and advertising agency TBWA/CHIAT/DAY to create the books.

Read more