I have been quite vocal about my belief that AI can become a very useful coworker in healthcare. Gene editing is a good example of why.

When a disease is caused by a genetic mutation, identifying the mutation is only the beginning. Researchers still need to determine the best place to intervene.

There can be multiple possible target sites. The editing system has its own constraints. Some parts of the genome are more accessible than others. And a guide that works well at one location may not work as well at another.

Then there is the risk of unintended edits.

For years, researchers have had to design candidate guide RNAs and test them experimentally. A lot of the work is essentially narrowing down which candidates are worth taking into the lab. This is where AI can act as a coworker.

Machine-learning models can learn from large datasets of previous gene-editing experiments and help predict which targets and guide RNAs are more likely to work, which may have higher off-target risk, and how genomic context can affect editing efficiency. Newer models are also using information about chromatin and the cellular environment, giving researchers more signals to consider before running the experiment.

And to clarify, the role of AI is to not work independently but to help the scientists focus their efforts on tests, experiments and deeper analysis. Instead of testing a much larger number of possibilities, researchers can use AI to narrow the search and focus their experimental work on the most promising candidates.

As gene editing moves from CRISPR cutting to more precise approaches such as base editing and prime editing, that ability to choose the right target becomes even more important. What makes me spend time reading about AI usecases in healthcare if mainly due to the potential that I see. We are talking about improving the quality of life, giving patients hope, something that no other usecase can even come close.

I hope in the coming years we will see AI working alongside the scientists, processing far more possibilities than a person can, and helping decide where to focus the next experiment. And in gene editing, even narrowing down where to look can have a huge impact.

I write about hashtag#artificialintelligence | hashtag#technology | hashtag#startups | hashtag#mentoring | hashtag#leadership | hashtag#financialindependence

PS: All views are personal