Summary: Researchers used AI to accelerate the discovery of treatments for Parkinson’s disease. They used machine learning techniques to screen millions of compounds and successfully identified five promising candidates that prevent the harmful accumulation of alpha-synuclein, a key protein in Parkinson’s pathology.
This AI-driven approach has accelerated the screening process tenfold and significantly reduced costs, potentially bringing new treatments to patients much sooner. The findings show a crucial advance in drug discovery.
Key facts:
- The AI method made it possible to identify compounds that inhibit the aggregation of alpha-synuclein, significantly accelerating a process that traditionally takes years.
- The improvement in screening efficiency is ten times faster and the cost reduction is a thousand times greater, marking a huge leap in drug discovery efficiency.
- More than six million people worldwide are affected by Parkinson’s and this number is expected to triple by 2040, underscoring the urgent need for effective treatments.
Fountain: University of Cambridge
Researchers have used artificial intelligence techniques to greatly accelerate the search for treatments for Parkinson’s disease.
Researchers at the University of Cambridge designed and used an artificial intelligence-based strategy to identify compounds that block the aggregation of alpha-synuclein, the protein that characterizes Parkinson’s.
The team used machine learning techniques to quickly screen a chemical library containing millions of entries and identified five very potent compounds for further investigation.
Parkinson’s affects more than six million people worldwide, and that number is expected to triple by 2040. There are currently no disease-modifying treatments available.
The process of screening large chemical libraries for drug candidates (which must be done long before potential treatments can be tested in patients) is time-consuming and expensive, and often unsuccessful.
Using machine learning, researchers were able to speed up the initial screening process tenfold and reduce the cost a thousandfold, which could mean potential treatments for Parkinson’s reach patients much faster.
The results are published in the magazine. Nature Chemistry Biology.
Parkinson’s is the fastest growing neurological disease worldwide. In the UK, one in every 37 people alive today will be diagnosed with Parkinson’s in their lifetime.
In addition to motor symptoms, Parkinson’s can also affect the gastrointestinal system, nervous system, sleep patterns, mood, and cognition, and may contribute to reduced quality of life and significant disability.
Proteins are responsible for important cellular processes, but when people have Parkinson’s, these proteins go rogue and cause nerve cells to die. When proteins misfold, they can form abnormal clumps called Lewy bodies, which build up inside brain cells, preventing them from functioning properly.
“One route to searching for potential treatments for Parkinson’s requires the identification of small molecules that can inhibit the aggregation of alpha-synuclein, which is a protein closely associated with the disease,” said Professor Michele Vendruscolo of the Department of Chemistry Yusuf Hamied, who led the investigation.
“But this is a time-consuming process: simply identifying a lead candidate for further testing can take months or even years.”
Although clinical trials are currently underway for Parkinson’s, no disease-modifying drugs have been approved, reflecting the inability to directly target the molecular species that cause the disease.
This has been a major obstacle in Parkinson’s research, due to the lack of methods to identify and collaborate with the correct molecular targets. This technological gap has severely hampered the development of effective treatments.
The Cambridge team developed a machine learning method in which chemical libraries containing millions of compounds are screened to identify small molecules that bind to amyloid aggregates and block their proliferation.
A small number of first-tier compounds were then experimentally tested to select the most potent aggregation inhibitors.
The information obtained from these experimental trials was fed back to the machine learning model iteratively, so that after a few iterations, very potent compounds were identified.
“Instead of doing screening experimentally, we do it computationally,” said Vendruscolo, co-director of the Center for Misfolding Diseases.
“By using the knowledge we gained from the initial screening with our machine learning model, we were able to train the model to identify the specific regions of these small molecules responsible for binding, then we can screen again and find more potent molecules.”
Using this method, the Cambridge team developed compounds to target pockets on the surfaces of aggregates, which are responsible for the exponential proliferation of the aggregates themselves. These compounds are hundreds of times more powerful and much cheaper to develop than those described above.
“Machine learning is having a real impact on the drug discovery process: it is 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, rather than just one. “So much is possible thanks to the huge reduction in both time and costs – it is an exciting time.”
Money: The research was carried out at the Health Chemistry Laboratory in Cambridge, which was established with support from the UK Partnership Research Investment Fund (UKRPIF) to promote the translation of academic research into clinical programmes.
About this news on AI research and Parkinson’s disease
Author: sara collins
Fountain: University of Cambridge
Contact: Sarah Collins – University of Cambridge
Image: Image is credited to Neuroscience News.
Original research: Open access.
“Discovery of potent inhibitors of α-synuclein aggregation through iterative structure-based learning” by Michele Vendruscolo et al. Nature Chemistry Biology
Abstract
Discovery of potent inhibitors of α-synuclein aggregation through iterative structure-based learning
Machine learning methods promise to reduce the costs and failure rates of conventional drug discovery processes.
This issue is especially pressing in the case of neurodegenerative diseases, where the development of disease-modifying drugs has been particularly challenging.
To address this issue, we describe here a machine learning approach to identify small molecule inhibitors of α-synuclein aggregation, a process implicated in Parkinson’s disease and other synucleinopathies.
Because proliferation of α-synuclein aggregates occurs via autocatalytic secondary nucleation, we aim to identify compounds that bind to catalytic sites on the surface of the aggregates.
To achieve this goal, we used structure-based machine learning iteratively to first identify and then progressively optimize secondary nucleation inhibitors.
Our results demonstrate that this approach leads to the easy identification of compounds two orders of magnitude more potent than those previously reported.
#revolutionizes #search #treatments #Parkinsons #Neuroscience #News