Artificial intelligence technology helped researchers identify a trio of chemicals that target faulty cells linked to age-related health issues, including certain cancers and Alzheimer’s disease.
The algorithm comb through a library of more than 4,300 chemical compounds and identified 21 drug candidates that could prompt cell senescence.
This is a phenomenon in which faulty cells stop multiplying but do not die off as they should and continue to release chemicals that can trigger inflammation.
Of those 21 candidates, the scientists zeroed in on three compounds – ginkgetin, oleandrin, and periplocin – which were able to remove defective cells without harming healthy ones when tested on human cells.
AI is increasingly becoming a fixture in medical and scientific research, able to sift through mountains of dense data far faster than a human ever could to aid in the diagnosis of and treatment for diseases.
![The research team headed up by experts at Scotland's University of Edinburgh devised an algorithm that successfully zeroed in on three natural compounds that can help stave off age-related health declines such as vision loss, loss of mobility, and Alzheimer's](https://i.dailymail.co.uk/1s/2023/06/14/21/72143351-12195871-The_research_team_headed_up_by_experts_at_Scotland_s_University_-m-100_1686775175470.jpg)
The research team headed up by experts at Scotland’s University of Edinburgh devised an algorithm that successfully zeroed in on three natural compounds that can help stave off age-related health declines such as vision loss, loss of mobility, and Alzheimer’s
The team led by researchers from the University of Edinburgh said their machine-learning technology led to several hundred-fold reductions in the cost of screening for effective ‘senolytic’ drugs.
‘Senolytic’ drugs induce death of senescent cells and improve health in humans.
Dr Diego Oyarzún, a co-author of the study, said: ‘This study demonstrates that AI can be incredibly effective in helping us identify new drug candidates, particularly at early stages of drug discovery and for diseases with complex biology or few known molecular targets.’
Cellular senescence occurs in a wide range of health conditions associated with getting older, including vision decline, osteoarthritis, a chronic lung disease known as Idiopathic pulmonary fibrosis, certain cancers, Alzheimer’s disease, and atherosclerosis, a cardiovascular condition that develops when sticky plaques build up in the arteries.
Cell senescence is a marker of an aging immune system.
It refers to a state of the cell cycle when cells cannot grow and replicate but don’t die either as part of a natural process known as apoptosis.
Instead, the senescent cells remain active within the body and release harmful substances that can cause inflammation and harm to nearby healthy cells, similar to how a moldy piece of melon would contaminate the entire fruit salad.
Because cell senescence entails a halt in the replication of cells, the process is believed to have a suppressive effect on tumors.
At the same time, senescence can promote cancer development by altering the environment around the cell.
The researchers said: ‘Besides their role in cancer and aging, the senescent program has been linked to adverse effects in a broad range of conditions, including osteoporosis, osteoarthritis, pulmonary fibrosis, SARS-CoV-2 infection, hepatic steatosis, and neurodegeneration.’
Only one therapy, a combination of Dasatinib plus Quercetin, has been shown in clinical trials to reduce the volume of senescent cells in mice.
Among the three compounds the researchers identified as able to eliminate senescent cells, oleandrin proved more potent than cardiac glycosides, the best-performing family of senolytic drugs.
The three compounds identified in the study, published in the journal Nature Communications, are natural products found in traditional herbal medicines.
Dr Vanessa Smer-Barreto, a co-author, said: ‘This work was borne out of an intensive collaboration between data scientists, chemists and biologists.
‘Harnessing the strengths of this interdisciplinary mix, we were able to build robust models and save screening costs by using only published data for model training. I hope this work will open new opportunities to accelerate the application of this exciting technology.’