Using Artificial Intelligence to Design Zinc Fingers: A New Approach to Gene Regulation and Therapeutic Development
“Engineered zinc finger proteins hold enormous promise for a variety of applications including as therapeutics for many genetic diseases. We combine artificial intelligence and biology to develop new technologies that will ultimately have great benefit for patients, in Canada and abroad.”
Zinc finger proteins are among the most common regulatory proteins in the human body. They bind to DNA or RNA and play essential roles in DNA repair, cell communication, and gene expression regulation—the process by which information from a gene is used to produce RNA molecules and proteins. Problems in these processes, including abnormal levels of protein production, can lead to disease. Because zinc fingers can recognize specific DNA sequences, they are strong candidates for gene editing and hold significant promise as therapeutic tools for numerous genetic diseases.
Dr. Philip M. Kim, a professor at the University of Toronto’s Donnelly Centre and the Departments of Computer Science and Molecular Genetics, has developed the ZFDesign AI technology. This technology can engineer zinc finger proteins by analyzing 50 billion potential zinc finger–DNA interactions to modify gene expression. This novel process combines artificial intelligence (AI) with high-throughput screening—methods that allow scientists to rapidly analyze many samples at the same time—to enable the precise design of zinc finger proteins for targeted DNA modification and potential disease treatment.
“While there had been much activity in engineering them for gene editing purposes, this largely stopped with the arrival of CRISPR-based technologies, which offered advantages. Now, in the AI era, zinc fingers are likely going to be the better technology for application as therapeutics,” says Dr. Kim.
Unlike CRISPR, which typically uses bacterial proteins to cut DNA at specific locations in the genome, zinc finger–based approaches can target precise regions to regulate gene expression—turning genes on or off without cutting DNA. Zinc fingers may also offer advantages such as reducing immune responses, natural compatibility with the human body, and better clinical delivery options due to their small size.
Dr. Kim’s research illustrates how combining AI and biology can lead to meaningful benefits for patients in Canada and beyond. The team believes this technology could enable a new generation of gene therapies across multiple disease areas and serve as a valuable research tool. “A number of untreatable diseases, many of them affecting young children and highly debilitating, may become treatable using this new technology,” Dr. Kim adds.
The work was funded by the Canadian Institutes of Health Research (CIHR), and the ZFDesign model was published in Nature Biotechnology in collaboration with Professors Timothy Hughes and Mikko Taipale from the University of Toronto and Marcus Noyes from the University of Pittsburgh.
Additional Resources
- Dr. Philip M. Kim – Donnelly Centre for Cellular and Biomolecular Research, University of Toronto
- Dr. Philip M. Kim’s Laboratory
- A universal deep-learning model for zinc finger design enables transcription factor reprogramming
- Date modified: