ZFDesign artificial intelligence program may speed development of gene therapy

"Zinc fingers" might prove key to treating complex genetically driven diseases. Zinc fingers are customizable proteins that can guide DNA repair by directing enzymes to cut faulty segments out of a person's DNA code, researchers said. However, the proteins are challenging to work with, because they form complex attachments to DNA. Researchers need to be able to tell how every zinc finger interacts with its neighbor for each desired genetic change, out of countless possible combinations.

To address this, researchers came up with ZFDesign, an artificial intelligence that uses a database of nearly 50 billion possible zinc finger-DNA interactions to help model and design gene edits. This technology could accelerate development of gene therapies for illnesses like cystic fibrosis, Tay-Sachs disease and sickle cell anemia, which are all caused by errors in the order of DNA letters that encode the operating instructions for every human cell, researchers said.

They said zinc finger editing offers a potentially safer alternative to CRISPR.

Zinc fingers are entirely human-derived, but CRISPR relies on bacterial proteins to interact with genetic code. These proteins could trigger a patient's immune system, which would identify them as foreign to the body and attack them as though they were a typical infection. Zinc finger tools also are smaller and might provide more flexible gene therapy techniques compared with CRISPR, researchers added.

"By speeding up zinc finger design coupled with their smaller size, our system paves the way for using these proteins to control multiple genes at the same time," said senior author Marcus Noyes, an assistant professor of biochemistry and molecular pharmacology at NYU Grossman School of Medicine in New York City. "In the future, this approach may help correct diseases that have multiple genetic causes, such as heart disease, obesity and many cases of autism," he said in the release. However, Noyes cautioned that zinc fingers are still difficult to control. They aren't always specific to a single gene, so some combos can lead to unintended changes in genetic code beyond the intended target. Because of this, the team next plans to refine the AI program so it builds more precise zinc finger groupings that prompt only the desired gene edit, Noyes said.

 

https://www.upi.com/Science_News/2023/01/26/artificial-intelligence-improve-gene-therapy/3961674750373/

 

کلمات کلیدی
//isti.ir/Z9Mv