Wednesday, March 29, 2023 | Back issues
Courthouse News Service Courthouse News Service

New computer model can identify cancer mutation cells, study finds

By identifying cancer cells that drive tumor growth, scientists hope to assist doctors in targeting such cells with anti-cancer drugs.

(CN) — In a study released Monday, researchers at MIT have developed a breakthrough computer model which will allow scientists to identify harmful cancer mutation cells, allowing them to better target such cells with anti-cancer drugs.

“We created a probabilistic, deep-learning method that allowed us to get a really accurate model of the number of passenger mutations that should exist anywhere in the genome,” said Maxwell Sherman, an MIT graduate student and one of the lead authors of the study. “Then we can look all across the genome for regions where you have an unexpected accumulation of mutations, which suggests that those are driver mutations.”

In the study published in the journal Nature Biotechnology, the scientists explained how the computer model “can rapidly scan the entire genome of cancer cells,” allowing them to identify which mutations in the cancer cells are possibly driving growth of tumors.

“The findings could help doctors to identify drugs that would have greater chance of successfully treating those patients,” according to a press release. 

The researchers said “at least 30% of cancer patients have no detectable driver mutation that can be used to guide treatment.”

While the human genome has been sequenced since 2003, scientists have struggled to determine the majority of mutations that can occur.

“There has really been a lack of computational tools that allow us to search for these driver mutations outside of protein-coding regions,” said Bonnie Berger, senior author of the study. “That's what we were trying to do here: design a computational method to let us look at not only the 2 percent of the genome that codes for proteins, but 100 percent of it.”

The sophisticated model looks at 37 types of cancer, allowing it to figure out mutation rates for each type.

“The really nice thing about our model is that you train it once for a given cancer type, and it learns the mutation rate everywhere across the genome simultaneously for that particular type of cancer,” Sherman said. “Then you can query the mutations that you see in a patient cohort against the number of mutations you should expect to see.”

Read the Top 8

Sign up for the Top 8, a roundup of the day's top stories delivered directly to your inbox Monday through Friday.