Machine-learning algorithm sheds light on DNA repair

By Folake Dosu  |  November 13, 2018

machine-learning-dna-repair

Mutations associated with diseases can manifest as insertions and deletions, depending on whether extra or missing DNA is involved. 

CRISPR-Cas9, an enzyme used for editing genes, is seen as a potential medical breakthrough, but scientists have struggled with applying the technology to a broad range of cell types in the body due to uncertainty about how these cells would repair the resulting breaks in DNA. 

ScienceDaily reports that a new study published in Nature describes how a machine learning algorithm is shedding light on how cells repair broken genes. 

According to the outlet, at the helm of this study were David Liu, the Richard Merkin Professor and director of the Merkin Institute of Transformative Technologies in Healthcare, and vice chair of the faculty at the Broad Institute; David Gifford, professor of computer science and biological engineering at MIT; and Richard Sherwood, an assistant professor of medicine in the Division of Genetics at Brigham and Women's Hospital.

"Machine learning offers new horizons for the development of human therapeutics. This study is an example of how combining computational experiment design and analysis with therapeutic goals can produce an unexpected therapeutic modality."

The algorithm in their study predicts cell responses to CRISPR-induced breaks in DNA, and results suggest this repair is more precise than had previously been thought, even sometimes repairing the gene to their intended form. Per ScienceDaily, researchers “successfully corrected mutations in cells taken from patients with one of two rare genetic disorders.”

This “genetic auto-correction” property of cells could enhance the CRISPR-based gene therapies expected to transform modern medicine.

"Machine learning offers new horizons for the development of human therapeutics," said Gifford to ScienceDaily, "This study is an example of how combining computational experiment design and analysis with therapeutic goals can produce an unexpected therapeutic modality."

"We don't currently have an efficient way to precisely correct many human disease mutations," Liu said to ScienceDaily. "Using machine learning, we've shown we can often correct those mutations predictably, by simply letting the cell repair itself."
 

Check out tech
innovation in
your backyard

Startup news + jobs

Hiring Resources

Recruitment 3.0—
stay informed.

Grow Your Company

Great Companies Need Great People. That's Where We Come In.

Recruit With Us