Unlocking the Fountain of Youth: AI Discovers Promising Anti-Ageing Medications

AI Finds Drugs that Could Fight Ageing and Age-Related Diseases

Published: July 6, 2023 2.20pm BST

Author: Vanessa Smer-Barreto

Research Fellow, Institute of Genetics and Molecular Medicine, The University of Edinburgh

Disclosure statement: Vanessa Smer-Barreto works for The University of Edinburgh. She receives funding from the Medical Research Council and The University of Edinburgh.

Partners: The University of Edinburgh provides funding as a member of The Conversation UK.

Email Twitter Facebook LinkedIn WhatsApp Messenger

Finding new drugs – called “drug discovery” – is an expensive and time-consuming task.

But a type of artificial intelligence called machine learning can massively accelerate the process and do the job for a fraction of the price. Researchers at The University of Edinburgh and the Spanish National Research Council have recently employed this technology to find three promising candidates for senolytic drugs – drugs that slow ageing and prevent age-related diseases.

Zombie Cells and Ageing

Senolytics work by targeting senescent cells, which are cells that are metabolically active but unable to replicate. These cells often accumulate in the body over time due to various environmental factors such as UV rays and chemical exposure. While preventing their replication stops the spread of DNA damage, senescent cells also release inflammatory proteins that can harm neighboring cells.

Elevated levels of senescent cells have been linked to a range of diseases, including type 2 diabetes, COVID-19, pulmonary fibrosis, osteoarthritis, and cancer. Studies in lab mice have shown that eliminating these cells using senolytics can ameliorate these diseases by selectively killing off the senescent cells while keeping healthy cells alive.

Currently, only two senolytic drugs, dasatinib and quercetin, have been tested in humans out of the known 80 senolytics. However, the traditional drug discovery process can take up to two decades and cost billions of dollars, making it crucial to find faster and more cost-effective methods.

AI in Drug Discovery

In their study, the researchers aimed to train machine learning models to identify new senolytic drug candidates. They fed the models with examples of known senolytics and non-senolytics, allowing the models to learn how to distinguish between the two and predict whether unknown molecules could also be senolytics.

Through a rigorous testing process, the researchers determined the best-performing model and tasked it with predicting the senolytic potential of 4,340 molecules. In a matter of minutes, the AI model delivered a list of 21 top-scoring molecules that it deemed to have a high likelihood of being senolytics.

In comparison, testing these 4,340 molecules in a lab would have taken several weeks and cost a significant amount of money. Furthermore, the researchers tested the top-scoring molecules on healthy and senescent cells and identified three, periplocin, oleandrin, and ginkgetin, that effectively eliminated senescent cells while sparing most of the normal cells.

The Potential of Interdisciplinary Collaboration

These promising findings highlight the potential of an interdisciplinary approach involving data scientists, chemists, and biologists. With access to high-quality data, AI models can accelerate the work of chemists and biologists in discovering new treatments and cures for diseases, particularly those with unmet needs.

The researchers have moved forward with further testing of the three candidate senolytics in human lung tissue. They hope to report their next results in two years’ time.

The discovery of new senolytic drugs through AI-driven drug discovery holds great promise in the fight against ageing and age-related diseases. It offers hope for developing more effective treatments in a shorter timeframe without breaking the bank.

Editorial and Advice

This groundbreaking research in AI-driven drug discovery provides a glimmer of hope for those seeking ways to combat ageing and age-related diseases. The ability of machine learning models to quickly identify potential senolytic drug candidates has the potential to revolutionize the drug discovery process.

However, it is essential to exercise caution and ensure that rigorous testing and validation of these candidates are conducted before introducing them into medical practice. The researchers have demonstrated their commitment to this by conducting further testing on human lung tissue. These thorough evaluations will ultimately determine the safety and efficacy of the identified senolytic drugs.

Furthermore, the interdisciplinary collaboration between data scientists, chemists, and biologists showcased in this study highlights the importance of multidisciplinary research in finding innovative solutions to complex problems. As we continue to explore the possibilities of AI in healthcare, fostering collaboration between different fields of expertise becomes even more critical.

In conclusion, while we eagerly await the results of the ongoing research, it is clear that AI-driven drug discovery has the potential to accelerate the development of new treatments for age-related diseases. The integration of AI into the drug discovery process has the power to provide life-changing therapies more efficiently, ultimately improving the quality of life for individuals worldwide.