Scientists studying the potential of machine learning approaches in drug discovery for Parkinson's disease and other neurodegenerative diseases are focusing on misfolded proteins that are a hallmark of such conditions. found that they could identify compounds that were “two orders of magnitude more potent” than those previously reported. , according to a new study.
Using this method, researchers in the UK and US were able to identify a compound that can effectively block the aggregation, or aggregation, of the alpha-synuclein protein, which is the root cause of Parkinson's disease, according to a study. .
“We anticipate that using machine learning approaches of the type described here could have significant benefits for researchers working in the field of protein misfolding diseases.” [such as Parkinson’s]and certainly early-stage drug discovery research in general,” the researchers wrote.
Their research, “Discovery of potent inhibitors of α-synuclein aggregation using structure-based iterative learning” was published in a magazine Natural chemical biology.
Machine learning is an approach to “speed up the entire process” of drug discovery
Parkinson's disease is characterized by the toxic accumulation of misfolded alpha-synuclein proteins within dopaminergic neurons (neurons responsible for releasing the neurotransmitter dopamine). Dopamine is a signaling molecule that plays a role in controlling movement. Parkinson's disease is caused by the progressive loss of these cells.
Despite efforts to identify compounds that block this toxic buildup, to date, there are no disease-modifying treatments available for Parkinson's disease.
Traditional strategies for identifying new treatments, which involve screening large chemical libraries for potential candidates before human testing, are slow, expensive, and often fail. there is.
In the case of Parkinson's disease, the development of effective treatments has been hampered by the lack of methods to identify appropriate molecular targets.
“One approach to searching for potential treatments for Parkinson's disease involves the identification of small molecules that can inhibit alpha-synuclein aggregation. …However, this is a very time-consuming process and further “Just identifying good candidates for testing can take months, even years,” Michele Vendruscolo, a professor at the University of Cambridge and lead author of the study, said in a university press release. Stated.
Now, researchers have developed a method that uses machine learning to rapidly screen chemical libraries containing literally millions of compounds. The goal was to identify small molecules that can block α-synuclein aggregation.
From a list of small molecules predicted to bind well to alpha-synuclein aggregates, the researchers selected a small number of top-ranking compounds and experimentally tested them as potent inhibitors of aggregation.
The results of these experimental assays were fed into a machine learning model to identify those with the most promising effects. This process was repeated several times and very potent compounds were identified.
“Instead of screening experimentally, we screen computationally,” Vendruscolo said.
Machine learning is having a major impact on the drug discovery process, speeding up the entire process of identifying the most promising candidates. For us, this means we can start working on multiple drug discovery programs instead of just one.
“By using the knowledge gained from the initial screen in a machine learning model, we were able to train the model to identify specific regions on these small molecules that are involved in binding. We then rescreened “We can find more powerful molecules,” Vendruscolo said.
Using this method, the researchers optimized an initial compound to target a pocket on the surface of alpha-synuclein aggregates.
In clinical tests using brain tissue samples from patients with Lewy body dementia (LBD) and multiple system atrophy (MSA), two forms of atypical parkinsonism, these compounds inhibited alpha-synuclein aggregation. was prevented.
“Machine learning is having a real impact on the drug discovery process, speeding up the entire process of identifying the most promising candidates,” Vendruscolo said. “For us, this means we can start working on multiple drug discovery programs instead of just one.”
Vendruscolo says, “Many things are now possible with significant savings in time and cost. These are exciting times.”
