A University of Zürich research team has developed a powerful new method so as to precisely edit DNA by way of combining cutting-edge genetic engineering along with artificial intelligence. This kind of technique opens the doors to more precise modelling when it comes to human diseases as well as lays the foundation for new generation gene therapy.
Apparently, accurate and targeted DNA editing by small point mutations along with integration of a whole gene through CRISPR/Cas technology has great potential for applications within biotechnology as well as gene therapy. But it is very significant that the so-called gene scissors do not cause any kind of unintended genetic alterations, although they maintain the genomic integrity so as to avoid the unintended side effects. Usually, double-stranded breaks within DNA molecules are precisely repaired within humans and also other organisms. However, occasionally this DNA end-joining repair happens to result in genetic errors.
Gene editing with immensely enhanced accuracy
It is well to be noted that now the scientists from the University of Zürich (UZH), Ghent University (Belgium), and the ETH Zürich have gone on to develop a new method that greatly enhances the precision when it comes to genome editing. By way of using artificial intelligence, a tool called Pythia anticipates how cells go on to repair the DNA after it gets cut by certain gene editing tools like CRISPR/Cas9.
The lead author, who pioneered the technology at UZH and is also currently a post-doc at Ghent University, Thomas Naert, said that their team developed tiny DNA repair templates, which act like molecular glue, and at the same time guide the cell to make accurate genetic changes.
Apparently, these AI-designed templates were first tested in human cell cultures, where they helped with highly precise edits as well as integrations. The approach was also validated across other organisms, such as Xenopus, which is a small tropical frog that is used in biomedical research, and also living mice, where the researchers successfully altered DNA within the brain cells.
Artificial intelligence can learn and predict DNA repair patterns
According to Thomas Naert, the repair follows patterns, and it is not random at all. And Pythia makes use of these patterns to their advantage. Traditionally when CRISPR goes ahead and cuts the DNA, scientists depend on the natural repair mechanisms of the cell to fix the break. While these repairs follow anticipated patterns, they can result in unwanted outcomes like destruction of the neighboring genes. Taking this into insight, the researchers simulated millions of possible editing outcomes by way of using machine learning, asking a simple but powerful question – what is the most efficient way in order to make a specific genome editing, given how the cell is most likely to repair itself?
Besides changing the individual letters of the genetic code or integrating an exogenously delivered gene, the method can also be made use of by fluorescent labeling of specific proteins. According to Naert, that is incredibly powerful, as it enables them to directly observe what every protein is doing in healthy as well as diseased tissue. Another benefit of this new method is that it goes on to work well across all the cells – even across organs having no cell division, like the brain.
Base for developing precise gene therapies
It is well to be noted that Pythia is named after the high priestess of the Oracle at the temple of Apollo of Delphi in antiquity, who was consulted in order to predict the future. In a similar way, this new tool enables scientists to anticipate the outcomes of gene editing with remarkable accuracy. According to the professor at the Institute of Anatomy of UZH as well as at ETH Zürich and also senior author of the study, Soeren Lienkamp, just as the meteorologists make use of AI so as to predict the weather, they are using it to forecast how the cells are going to respond to certain genetic interventions. This kind of productive power is necessary if they want to ensure gene editing is safe, dependable, and also clinically very useful.
Lienkamp goes on to add that what excites them the most is not just the technology itself, but also the possibility that it opens. Pythia goes on to bring together large-scale AI prediction along with real biological systems, right from culture cells to animals. This kind of tight loop between modeling as well as experimentation points is fast becoming very useful, for instance, when it comes to precise gene therapies.
Apparently, this work goes on to create new possibilities in order to understand genetic disease as well as develop gene therapy and also for neurological diseases that are both safer and more effective.
