Researchers have developed an Artificial Intelligence (AI) tool that can predict the efficacy of genome editing repair options, reducing errors in the correction of genetic diseases.
Genome editing technologies can provide effective options for treating genetic diseases. Methods such as CRISPR/Cas9 gene scissors can directly address the causes of genetic diseases in the DNA. These scissors are used in labs to make targeted modifications to the genetic material in cell lines and model organisms, as well as to study biological processes.
Researchers have developed a new CRISPR/Cas9 method called prime editing. Unlike typical gene scissors, which create a break in both strands of the DNA molecule, prime editing only cuts and repairs DNA on a single strand The prime editing guide RNA (pegRNA) can target the relevant site in the genome precisely, providing new genetic information. The information is then transcribed by a translation enzyme before being incorporated into the DNA.
Genome editing can cause unwanted side effects
Prime editing has the potential as a method of repairing dangerous mutations in patients’ genomes. However, it is important that genome editing is applied successfully in order to avoid unwanted side effects such as errors in DNA correction and alteration of DNA in the genome.
According to previous research, prime genome editing can lead to a significantly lower number of unintended changes than conventional CRISPR/Cas9 methods. Currently, researchers have to spend a long time optimising the pegRNA for a specific target in the genome.
“There are over 200 repair possibilities per mutation. In theory, we would have to test every single design option each time to find the most efficient and accurate pegRNA,” said Gerald Schwank, professor at the Institute of Pharmacology and Toxicology at the University of Zurich (UZH).
The UZH researchers wanted to find an easier solution to this issue. They worked alongside Michael Krauthammer, UZH professor at the Department of Quantitative Biomedicine, to develop a new method of predicting pegRNA efficiency.
The team assessed 100,000 different pegRNAs in human cells, allowing them to generate a comprehensive prime genome editing data set. This enabled the researchers to establish which properties of a pegRNA positively or negatively influence the genome editing process. They did this by examining the length of DNA sequences, the sequences of DNA building blocks and the shape of DNA molecules.
The researchers were able to develop an AI-based algorithm which could recognise relevant patterns in the pegRNA. Using these patterns, the AI tool can predict both the effectiveness and accuracy of genome editing with a particular pegRNA.
“In other words, the algorithm can determine the most efficient pegRNA for correcting a particular mutation,” explained Michael Krauthammer.
Prime editing could be use to treat hereditary diseases
The new technology successfully tested in human and mouse cells and is now freely available to researchers.
The researchers acknowledge that further pre-clinical studies are still required before the new prime genome editing tool can be administered in humans. However, the team are confident that it will be possible to use genome editing to repair the DNA mutations of hereditary diseases such as sickle cell anaemia, cystic fibrosis and metabolic diseases in the near future.
