Neural network has found new drugs to slow aging
In just five minutes, the neural network identified 21 substances that are highly likely to be drugs for slowing down aging. If such a search had been conducted in a conventional laboratory, the work would have lasted several weeks and cost about 60,000 euros.Today, artificial intelligence is transforming many fields, including medical research. It offers unprecedented opportunities to accelerate drug discovery, reduce costs and discover new ways to fight disease. Neural networks quickly analyze large amounts of data and identify patterns that are otherwise difficult to detect. This can speed up the work of scientists seeking to find cures for diseases, especially those for which there is no solution yet.
In this case, an international team of scientists used a neural network to identify promising candidates for so-called senolytic drugs that slow aging and prevent age-related diseases. A specially trained neural network predicted the potential efficacy of many thousands of molecules, reducing the time and costs associated with traditional laboratory research.
Twenty-one compounds were identified first, and then of those, the three most promising were periplocin, oleandrin and ginkgetin. They could destroy senescent cells while preserving the majority of normal cells. More detailed tests have shown that oleandrin, for example, is more effective than the best-known senolytics in its class today. The three candidate senolytics identified by the neural network are now being tested on human tissue. The study is published in the journal The Conversation.