Parkinson's disease, the swiftest spreading neurological condition globally, impacts over six million individuals. It's not just a movement disorder, and it also implicates the gastrointestinal tract, the nervous system, sleep cycles, and can influence mood and cognitive functions, ultimately diminishing the sufferer's life quality and ability to function independently.
Existing treatments are primarily aimed at alleviating symptoms, without addressing the disease's progression. Hence, the development of substances that can thwart αS aggregation is of paramount importance in the quest for more effective Parkinson's therapies. Scholars from the University of Cambridge have penned a study unveiling a structure-based, iterative learning strategy aimed at unearthing powerful inhibitors that can prevent the clumping of α-synuclein (αS). This protein's accumulation is intimately linked to the neurological decay characteristic of Parkinson's.
Fig. 1. An iterative learning approach to identify compounds with therapeutic potential for PD. (Horne R I.; et al. 2024)
The process of screening large chemical libraries for drug candidates is time-consuming and expensive, and often unsuccessful. A major obstacle to Parkinson's research is the lack of methods to identify and interact with the correct molecular targets.
The researchers designed and used an artificial intelligence (AI)-based machine learning method to screen a chemical library containing millions of compounds to identify compounds that block the aggregation or polymerization of αS, a protein characteristic of PD. The machine learning approach described in the article includes three main components: the experimental data, the variational autoencoder and the model for training and prediction.
The researchers performed experimental tests on a small number of top-ranked compounds that were screened to select the most effective inhibitors of αS aggregation. The information gained from these experimental analyses is fed into the machine learning model in an iterative manner, so that after a few iterations, the most potent compounds are identified.
The compounds identified through the machine learning method exhibited higher activity and lower inhibitory concentrations than the known clinical trial molecule Anle-138b in inhibiting αS aggregation. The binding affinity of these compounds for αS was also verified under different pH conditions and had no effect on Aβ42 aggregation, showing specificity for αS.
In addition, these compounds also showed satisfactory results in αS aggregation inhibition experiments with brain-derived seeds, suggesting that they may also be effective against αS aggregation in the brain tissue of PD patients. These findings provide new potential drug candidate molecules for the treatment of PD and demonstrate the potential for the application of machine learning in drug discovery.
The iterative machine learning strategy described in this article provided a higher optimization rate than traditional high-throughput screening in the discovery of αS aggregation inhibitors and could discover new compounds with significant differences from the parent structure. This means that potential treatments for Parkinson's can reach patients faster.
Our company understands the importance of AI approaches such as machine learning and deep learning in today's drug development landscape. We are focused on utilizing AI-based technologies to accelerate the entire process of identifying the most promising drug candidates. Partner with us to shorten drug development time and reduce the cost and failure rate of the traditional drug development process. If you are interested in our services or have a question, please feel free to contact us for more details.
Original Article:
Horne R I.; et al. (2024). Discovery of potent inhibitors of α-synuclein aggregation using structure-based iterative learning. Nature Chemical Biology. 2024: 1-12.
Services Related in the Article:
Target Identification
AI-powered Drug Discovery and Design
AI-assisted Drug High Throughput Screening
We harness the power of artificial intelligence to transform the landscape of drug discovery. Our mission is to accelerate the development of life-saving medicines by leveraging cutting-edge AI technologies.
