Researchers led by the University of Zurich have unveiled a novel method that leverages artificial intelligence to improve the precision of gene editing. This innovation promises to revolutionize disease modeling and pave the way for safer, more effective gene therapies.
A team of scientists from the University of Zurich (UZH), in collaboration with Ghent University and ETH Zurich, has achieved a significant breakthrough in the world of genetic engineering. Their innovative technique, which blends artificial intelligence with CRISPR/Cas9 technology, takes DNA editing precision to new heights.
This development, published in the journal Nature Biotechnology, is poised to transform the modeling of human diseases and lay a sturdy foundation for next-generation gene therapies.
Currently, CRISPR/Cas technology shows great potential for applications in biotechnology and gene therapy due to its ability to make precise, targeted edits to DNA. However, ensuring that these “gene scissors” do not cause unintended genetic mutations is crucial to maintaining genomic integrity and preventing adverse side effects.
The newly developed method, spearheaded by lead author Thomas Naert, who was at UZH at the time and is now a postdoctoral researcher at Ghent University, utilizes a tool called “Pythia.” This AI-driven system predicts how cells will repair their DNA following gene editing, thereby enabling more precise interventions.
“Our team developed tiny DNA repair templates, which act like molecular glue and guide the cell to make precise genetic changes,” Naert said in a news release.
These AI-designed templates were tested in human cell cultures, where they yielded highly accurate gene edits and integrations.
The researchers also validated the approach in other organisms, including the model organism Xenopus, a small tropical frog widely used in biomedical research, and in living mice, where they successfully edited DNA in brain cells.
“DNA repair follows patterns; it is not random. And Pythia uses these patterns to our advantage,” added Naert.
Traditional CRISPR methods rely on the cell’s natural repair mechanisms, which can sometimes lead to unintended genetic alterations. By leveraging machine learning, the researchers simulated millions of potential editing outcomes to determine the most efficient way to make specific genetic changes.
In addition to editing the genome, the method also enables fluorescent labeling of particular proteins.
“That is incredibly powerful,” Naert added, “because it allows us to directly observe what individual proteins are doing in healthy and diseased tissue.”
This versatility extends to all cell types, even those in non-dividing organs like the brain.
Pythia is aptly named after the high priestess of the oracle at the Temple of Apollo at Delphi, who was famed for her prophetic capabilities. In a similar vein, this new tool allows researchers to forecast the outcomes of gene editing with extraordinary precision.
“Just as meteorologists use AI to predict the weather, we are using it to forecast how cells will respond to genetic interventions. That kind of predictive power is essential if we want gene editing to be safe, reliable and clinically useful,” added senior author Soeren Lienkamp, a professor at UZH and ETH Zurich.
“What excites us most is not only the technology itself, but also the possibilities it opens. Pythia brings together large-scale AI prediction with real biological systems. From cultured cells to whole animals, this tight loop between modeling and experimentation points is becoming increasingly useful, for example in precise gene therapies,” Lienkamp added.
The implications of this work are profound, ranging from a deeper understanding of genetic diseases to the development of gene therapies for neurological conditions and other illnesses. With enhanced safety and efficacy, this advanced gene editing method stands to make a significant impact on the fields of biomedical research and therapeutic medicine.
Source: University of Zurich