Researchers at the University of California, San Diego and IBM, have teamed up to develop an interesting artificial intelligence (AI) tool. It can estimate a person’s age based on the composition of the gut, skin and oral microbes. It comes after many researchers delved into the interesting link between aging and billions of microbes in the body. Now we’re getting new applications based on these research results.
(From: UCSD, via New Atlas)
Last year, the Singapore team conducted an intestinal microbiome transplant trial between elderly and young mice. The results suggest that this may alter bacterial populations to balance many of the systemic defects caused by the aging of organisms.
To investigate whether human microbes can be used for age indicators, the new study draws on the power of modern machine learning techniques. The team explored three different human microbiomes, covering the skin, mouth and intestines.
Data were then collected from 10 previous studies and nearly 9,000 microbiomic samples were taken from subjects aged 18 to 90.
Interestingly, the modeling found that the time-arranged characteristics of the skin microbiome can be used as the best indicator of accurate prediction (estimated errors within 3.8 years).
Skin microbiomes are sampled from the hands or forehead by swabs, and studies have shown that two locations provide equally accurate predictions that have little to do with population and geographic location.
The accuracy of oral microbiome is close to that, with an error of 4.5 years. The second is the gut microbiome, with an error of about 11.5 years.
The researchers speculated that the predicted accuracy of the skin microbiome was related to general physiological changes in the skin and age.
As we age, most people’s skin becomes drier and reduces serum production, and these changes are clearly reflected in the components of the surface microbiome.
It should be noted that the current results are based on microbiomic component data from subjects from the United States, China, the United Kingdom and Tanzania, and that greater and in-depth research is needed to better understand e-species and geographic differences.
Details of the study have been published in the recent issue of the journal mSystems. Originally published as:
Human Skin, Oral, and Gut Microbiomes PredictAge