The primary medical potential of the gene-editing technology CRISPR is to eliminate many types of genetic diseases,. This could be achieved through eliminating certain genetic codes responsible for specific illness. There are various barriers to achieving this, ranging from public policy, to funding, and the current state of the technology itself.
Another issues arises with so-called off-target effects. This means induced mutations at sites other than the intended on-target site. This is of particular concern with therapeutic and clinical applications. These issues, for example, have been raised by Dr. J. Keith Joung, of Massachusetts General Hospital, at the American Society of Hematology’s recent workshop on genome-editing.
To help address off-target effects, the technology giant Microsoft is hoping to use artificial intelligence to help to fix this problem. Unintended changes could, for example, inadvertently result in new health problems for a patient, such as cancer.
CRISPR (Clustered regularly-interspaced short palindromic repeats), as Digital Journal has reported (see: “Is CRISPR technology set to change biological science?”), is a biological cut-and-paste technique that allows researchers detect a gene defect within living cells and then use molecular “scissors” to make changes. Changes include deleting the gene; repairing it; or replacing it.
As an example of what can be achieved, Jennifer Doudna and Emmanuelle Charpentier repurposed a protein called Cas9 protein to develop a low-cost, precise and straightforward gene editor for initial studies, using an animal model to address a severe type of muscular dystrophy affecting males called Duchenne muscular dystrophy. Symptoms of this condition include muscle weakness, an doss of intellectual function.
To make CRISPR more accurate researchers around the world are attempting to fine-tune CRISPR. Microsoft is of the view that artificial intelligence can assist with this and the company has developed a new tool called Elevation. The aim of the platform is to predict off-target effects when scientists edit genes with CRISPR technology. Here, according to Gizmodo, Elevation can suggest which approach is less likely to result in off-target effects for a particular gene. The platform will improve in time through machine learning.
The application of Elevation has been described in the journal Nature Biomedical Engineering, with the paper titled “Prediction of off-target activities for the end-to-end design of CRISPR guide RNAs.”
