Researchers at the University of Cambridge in the UK have significantly accelerated the development of treatments for Parkinson's disease using artificial intelligence (AI) technology. They devised and employed an AI-based strategy to identify small molecules that inhibit the aggregation of α-synuclein, a characteristic protein of Parkinson's disease. The findings were published in the latest issue of Nature Chemical Biology.
Parkinson's disease affects over 6 million people worldwide, and this number is expected to double by 2040. When people develop Parkinson's disease, certain proteins misbehave, leading to the death of nerve cells. When these proteins misfold, they can form abnormal aggregates called Lewy bodies, which accumulate inside brain cells and disrupt their normal function.
One approach to finding potential treatments for Parkinson's disease is to identify small molecules that can inhibit the aggregation of α-synuclein. In this study, the team used machine learning techniques to rapidly screen a chemical library containing millions of entries to identify molecules that bind to and prevent the proliferation of amyloid-like protein aggregates. Ultimately, they identified five highly effective compounds.
By leveraging AI technology, researchers accelerated the initial screening process tenfold and reduced costs by a factor of a thousand, meaning the development of potential therapies for Parkinson's disease can proceed much faster.
Using this approach, the research team developed a compound that targets pockets on the surface of aggregates, which are responsible for the exponential growth of the aggregates themselves. This compound's potency is hundreds of times greater than previously reported, and its development costs are much lower.
Professor Michele Vendruscolo, who led the study, stated that machine learning is having an impact on the drug discovery process by expediting the identification of the most promising drug candidates. With significantly reduced time and costs, multiple drug development programs can be pursued in the future.
