AI discovers potential motor neurone disease treatment
Image credit: Dreamstime
University of Sheffield researchers working with artificial intelligence (AI) tools to search for potential treatments for motor neurone disease (MND) have identified a candidate drug which has proved effective in preclinical trials.
MND, a rare condition also known as amyotrophic lateral sclerosis (ALS), attacks the motor neurones in the brain and spinal cord necessary for muscle control, causing weakness, wasting, and leading to almost complete paralysis. There are no known treatments to cure the disease, and just two FDA-approved drugs to slow its progression.
Most famously, theoretical physicist Stephen Hawking was diagnosed with MND as a young man, and has outlived his short life expectancy by decades.
“Many doctors call it the worst disease in medicine and the unmet need is huge,” said Dr Richard Mead, a researcher at the Sheffield Institute of Translational Neuroscience who has adopted the use of AI in his medical research.
The introduction of AI to medicine is accelerating the pace of research by analysing databases with inhuman speed to identify new targets and drugs. MND is a condition in which AI could prove a particularly effective tool due to the explosion in the amount of genetic information available about the condition.
Dr Mead’s Sheffield team, working with Benevolent AI, have found that one of the candidate drugs proposed by their AI tools has produced promising results, being capable of protecting motor neurone cells and potentially delaying the onset of MND in preclinical experiments.
“What we are trying to do is find relationships that will give us new targets in disease,” said Dr Jackie Hunter, CEO of Benevolent AI, who believes that the future of drug discovery could require AI to be sustainable. “We can do things so much more dynamically and be really responsive to what essentially the information is telling us.”
The Sheffield researchers are now discussing plans for moving their treatment to human trials.
Previously, researchers at the Barrow Neurological Institute in Arizona used IBM’s cognitive system, Watson, to discover five new genes linked to MND. The researchers credited Watson with reducing the discovery time from years to months.
Motivated by early success stories of using AI tools to accelerate drug discovery, a number of start-ups have begun to appear focused on AI in pharmacological research.
Meanwhile, researchers at Carnegie Mellon University and the Hebrew University of Jerusalem have developed an AI-based method for mining databases of research papers and patents to identify possibilities for new inventions beyond healthcare.