Aging: new intelligent learning system shows strong natural product screening ability

The census bureau predicts that by 2030, one in five people in the United States alone will be over 65 years old Many of today's most prevalent, costly and devastating diseases, including cancer, cardiovascular disease, Alzheimer's disease and type II diabetes, will become the major killers of the elderly In order to prevent this situation in the long run, it is urgent to develop an anti-aging drug or other protective interventions to protect the health of the elderly, reduce the incidence of disease and maintain productivity One strategy to speed up this process is to readjust the existing FDA approved drugs, which have the potential of anti-tumor and anti-aging The first two drugs are metformin and rapamycin Although rapamycin and metformin are different in clinical use, they are both inhibitors of mTOR, and have many benefits of anti-aging, anti-cancer and anti cardiovascular diseases Rapamycin is an immunosuppressant used after kidney transplantation, but it can also prolong life in many animal models, including yeast, Drosophila and mice Rapamycin has been shown to reduce the risk of cancer after kidney transplantation Like rapamycin, metformin is also an inhibitor of mTOR, indirectly through a variety of mechanisms Metformin is the most famous biguanide drug, which is the first-line treatment of type II diabetes and metabolic syndrome Its role seems to be multifaceted and beneficial for the treatment of many age-related diseases, including cancer and cardiovascular diseases Working flow chart (source: Aging) researchers from several universities and research institutions in the United States, the United Kingdom, Canada and Russia published a landmark study on the identification of metformin and rapamycin natural analogues in the journal aging In this study, researchers used in-depth learning neural network to describe the safety, gene and pathway level similarity of more than 800 natural compounds to metformin and rapamycin, so as to determine the anti-cancer and anti-aging effects of these natural compound antibodies, and understand the related side effects DL based similarity research (source: Aging) earlier this year, a comprehensive system of young AI emerged, which can use the latest in-depth learning to track various aging biomarkers One of the goals of this system is to help determine the molecular combination that can achieve the desired effect The team's analysis identified a number of new candidates for metformin and rapamycin mimics that had never been reported before In particular, they identified allantoin and ginsenoside as metformin simulants, epigallocatechin gallate and isoliquiritigenin as simulants of rapamycin, and withaferin a as strong simulants of both In addition, their analysis also identified four previously undeveloped natural compounds, which are relatively strong mimics of rapamycin Genetic and pathway level similarity studies (source: Aging) due to the in-depth classification of safety profiles related to these compounds, the new candidate simulants identified in the study may have less side effects than metformin and rapamycin, of course, the actual effect needs to be further verified by clinical tests This study is not only of great significance for the identification of new metformin and rapamycin candidate analogues, but also shows a particularly powerful screening method, which can be used for the screening of other known drugs and compounds This kind of AI intelligent system has a special prospect for the future experimental verification work Paper link: http://www.aging-us.com/article/101319/text