Researchers have used machine learning techniques to significantly accelerate the search for treatments for Parkinson's disease.
Researchers at the University of Cambridge designed and used a machine learning-based strategy to identify compounds that block aggregation, or aggregation, of alpha-synuclein, a protein that characterizes Parkinson's disease.
This technology was used to rapidly screen a chemical library containing millions of entries and identify five highly potent compounds for further investigation.
The study, “Discovery of potent inhibitors of α-synuclein aggregation using structure-based iterative learning,” natural chemical biology.
Speeding up the screening process for the onset of Parkinson's disease
Parkinson's disease affects more than 6 million people worldwide, and that number is expected to triple by 2040. However, no disease-modifying treatments for Parkinson's disease are currently available.
This is a major hurdle in Parkinson's disease research, as we lack the means to identify and address the correct molecular targets.
This technological gap significantly hinders the development of effective treatments.
Screening drug candidates from large chemical libraries must be done well before testing potential treatments on patients, is extremely time consuming, expensive, and often unsuccessful.
By using machine learning, researchers were able to speed up the initial screening process by a factor of 10 and cut costs by a factor of 1,000. This could mean potential Parkinson's disease drugs could reach patients much faster.
“One way to explore treatments for Parkinson's disease requires the identification of small molecules that can inhibit the aggregation of alpha-synuclein, a protein closely associated with the disease,” said Yusuf, who led the study.・Professor Michel Vendruskoro of the Hameed Department of Chemistry said: the study.
“However, this is a very time-consuming process, and it can take months or even years just to identify good candidates for further testing.”
How machine learning techniques were tested
The Cambridge team developed a machine learning method that screens chemical libraries containing millions of compounds to identify small molecules that bind to amyloid aggregates and block their growth.
We then experimentally tested a small number of the top compounds to select the most potent aggregation inhibitors.
The information obtained from these experimental assays was iteratively fed back into the machine learning model, and after several iterations, highly potent compounds were identified.
“Instead of screening experimentally, we screen computationally,” Vendruscolo said.
“Using the knowledge gained from the initial screening with the machine learning model, we were able to train the model to identify specific regions on these small molecules that are involved in binding. We can rescreen and find more powerful molecules to treat the disease.”
She added: “Machine learning is having a real 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.”
