University of Zurich Combines Gene Scissors With AI

(CONNECT) An international team of researchers led by the University of Zurich (UZH) has made the gene editing tool CRISPR/Cas9 even more precise. After a CRISPR intervention, a cell usually rejoins the two loose ends of a DNA precisely by itself. However, mistakes are occasionally made, possibly with negative consequences.

The newly developed artificial intelligence (AI) called Pythia has been trained with millions of possible gene editing results and can now answer this "simple but important question", according to a UZH report: What is the most efficient way to make a particular small change in the genome, given how the cell is likely to repair itself?

"DNA repair follows patterns; it is not random," explains Thomas Naert, first author of the study. "And Pythia uses these patterns to our advantage." It can therefore learn and predict DNA repair patterns. According to the article, Naert pioneered this technology at UZH and is currently conducting postdoctoral research at the University of Ghent.

The repair templates designed by AI were initially tested in human cell cultures, where they enabled the high-precision editing and integration of genes. The approach was also validated in living small frogs and in mouse brain cells.

"In addition to the technology itself, we are also excited about the possibilities it opens up," says last author Soeren Lienkemp, Professor at the Institute of Anatomy at UZH and the Swiss Federal Institute of Technology in Zurich (ETH). "Pythia brings together AI predictions based on large amounts of data with real biological systems." This is becoming increasingly useful for precise gene therapies in particular. According to the UZH, this work creates new opportunities for understanding genetic and neurological diseases and developing gene therapies that are both safer and more effective. ce/mm