Thanks to AI, researchers found new anti-aging medicines

Many significant advancements in the past year have been propelled by artificial intelligence. But while super-intelligent chatbots and rapid art generation have taken over the internet, AI has even gone so far as to take on one of the major issues facing humanity: aging.

According to this article, machine-learning systems have recently been employed in the field of drug discovery, thanks to research from the University of Edinburgh, which has led to the identification of a number of new anti-aging medicines.

Machine learning is concerned with using data to simulate human learning and improving accuracy as more data is put into it. But, this particular algorithm was hunting for a new senolytic medicine. In the past, it has been used to generate chess-playing robots, self-driving cars, and even on-demand TV suggestions.

Senolytics are essentially a class of medication that slows the aging process and guards against age-related illnesses. They function by eliminating senescent cells, which are damaged cells that can emit inflammatory compounds despite being unable to reproduce.

Senolytics are potent medications, but their development can be costly and time-consuming. Vanessa Smer-Barreto, a research fellow for the University of Edinburgh’s Institute of Genetics and Molecular Medicine, became aware of this and resorted to machine learning.

“Generating your own biological data can be really expensive, and it can take up a lot of time, even just to gather training data”, she explains.

“What made our approach different to others is that we tried to do it on limited funds. We took training data from existing literature and looked into how to utilize this with machine learning to speed things up”.

She discovered three viable alternatives for these kinds of drugs by employing a machine-learning algorithm.

Smer-Barreto and her colleagues accomplished this by teaching an AI model to differentiate between known senolytics and non-senolytics by feeding the model samples of each. Based on how closely the new molecules matched the pre-fed examples, they could then employ this information to determine whether they were senolytics.

Only two of the approximately 80 senolytics that are known have been tried on people. Even though that seems like a small percentage, drugs take 10 to 20 years and a lot of money to reach the market.

The scientists reviewed a wide range of articles, but they were selective in their analysis, focusing on only 58 chemicals. This allowed them to eliminate any compounds whose outcomes were unclear.

The machine-learning model received 4,340 molecules in all, and in just five minutes, it produced a list of results. The model determined that 21 of the highest-scoring compounds were most likely to be senolytics. Without the machine-learning model, this procedure may cost a lot of money and take weeks to complete. Lastly, two different cell types—healthy and aging—were used to examine the possible medication possibilities.

Three of the top 21 scoring compounds were able to kill aging cells while maintaining the viability of healthy cells. To learn more about how these novel senolytics interact with the body, more testing was conducted on them.

Despite the study’s success, this is only the beginning of the investigation. The next step, according to Smer-Baretto, is to work with clinicians at her university to evaluate the medications found in their samples of robust human lung tissue.

The team wants to examine whether they can slow down the aging process in the tissue of injured organs. Smer-Baretto emphasizes that, especially early on, the patient won’t always receive a large dose of a medication. These medications, which may be given locally or in tiny doses, are also being studied first on tissue models.

“It is essential that with any drug that we are administering or experimenting with, we consider the fact that it may do more harm than good”, says Smer-Baretto.

“The drugs have to go through many stages first, and even if they make it through to the market, it will have gone through a host of safety concerns tests first”.

There is nothing blocking AI from being used in other fields, even though this kind of data analysis was used on pharmaceuticals connected to aging.

“We had a very specific approach with the data, but there is nothing stopping us from applying similar techniques towards other diseases, such as cancer. We’re keen to explore all avenues”.

AI is changing the way we approach creativity, but medicine will also be a field that will be deeply influenced by it, especially in the discovery of new drugs but also in care and diagnostics that can be increasingly personalized to the patient’s needs and accelerated through preliminary diagnoses made by AIs.