AI is developing drugs for aging.

Artificial intelligence has become a driving force behind many large-scale developments over the past year. It has now gone far beyond just creating images and texts. Scientists have engaged artificial intelligence in the fight against aging – one of humanity’s greatest challenges. Machine learning systems have been used to discover new drugs targeting the human biological clock.

AI reduces the time needed to search for anti-aging drugs.

As is known, the more data a model has, the higher the accuracy of its actions. Previously, artificial intelligence was used to create robots that play chess, self-driving cars, and even on-demand television recommendations. This time, scientists have employed a specific algorithm to search for new senolytic drugs.

Senolytics are a form of medication that can slow down aging and prevent age-related diseases. They eliminate senescent cells—damaged and unable to reproduce, but capable of releasing substances that cause inflammation.

Despite the fact that senolytics are powerful drugs, their development can be expensive and time-consuming. Noticing this, Vanessa Smer-Barreto, a research fellow at the Scottish Institute of Genetics and Molecular Medicine, turned to artificial intelligence.

According to her, creating one’s own biological data can be a very expensive process that requires a lot of time, but researchers have been trying to solve this problem with limited funds. Ms. Smer-Barreto says they took training data from existing literature and considered how to use it with AI to speed up the work.

Using a machine learning algorithm, the researcher was able to identify three promising candidates for potential anti-aging drugs. To achieve this, Ms. Smer-Barreto and her colleagues provided the AI model with examples of known senolytics and non-senolytics, training the system to distinguish between them. This data could then be used to predict the presence of previously unknown molecules for combating cellular aging.

Just 5 minutes – and the cure for aging has been found.

Approximately 80 senolytics are known, but only 2 of them have been tested on humans. Along with enormous costs, it takes 10-20 years for anti-aging drugs to reach the market.

The Smer-Barreto team reviewed a wide range of documents but approached the results selectively. The researchers limited themselves to only 58 compounds, excluding any options that provided ambiguous results.

In total, 4,340 molecules were input into the AI model, which returned a list of results in just 5 minutes. The model identified 21 molecules with the highest scores that it believed were likely senolytics. Without AI, this process could have taken weeks and required huge amounts of money to achieve these results.

Finally, potential drug candidates were tested on two types of cells – healthy and aged. Out of 21 molecules, only 3 were able to eliminate aging cells while keeping normal cells alive. These new senolytics were then subjected to further testing to understand how they interact with the body.

One step closer to creating anti-aging drugs.

Although the research was successful, as noted by the publication Sciencefocus This is just the beginning. According to Ms. Smer-Barretto, the next step is to collaborate with clinicians and attempt to test the identified drugs on samples of healthy human lung tissue.

With the help of upcoming tests, the research team hopes to see if these drugs can combat the aging of damaged organ tissues. Ms. Smer-Barretto notes that it is not necessary to administer a large dose of the medication to the patient, especially in the early stages. Generally, the drugs will first be tested on tissue models, and only then will they be administered locally or in microdoses.

According to Ms. Smer-Barretto, the drugs must go through many stages initially. Even if they reach the market, they will first undergo a series of safety tests.

Despite the fact that this data research method has been used to search for anti-aging drugs, nothing prevents the application of AI in other areas. Similar models could be applied to drugs for other diseases (for example, cancer).