Any sufficiently complex set of data that changes with age can be used to produce an aging clock, given a database of measures from people of various ages. Machine learning is applied to discover algorithmic combinations of that data that predict age. This is thought to produce outcomes that reflect biological age; a person with a predicted age higher than chronological age has a greater burden of damage and dysfunction. No clock is fully understood, in the sense that it is unknown at the time of creation as to how exactly the clock will react to a higher or lower burden of any one specific form of cell and tissue damage or consequent dysfunction in aging. This makes clocks hard to use in the way that we would like to use them, to speed up the process of evaluating potential rejuvenation therapies by providing a rapid, low cost measure of the efficacy of a given treatment.
Biological age reflects the current state of the body, considering the aspects of lifestyle, environment, and hereditary component. Currently there is no universal formula for determining it, but there are markers that can be used to calculate it. This study aims to develop and compare two models for calculating biological age based on laboratory blood tests and composition of gut microbiota.
The biochemical model of biological age uses 7 indicators and is gender-specific (general - cystatin-C, IGF-1, DHEAS, only for females - homocysteine, urea, glucose, zonulin, only for males - HbA1c, NT-proBNP, free testosterone, hs-CRP). The microbial model requires the input of percentages of 45 bacterial species as indicators of the gut microbiota. Both methods demonstrate high predictive accuracy (mean absolute error ~ 6 years, R-squared > 0.8) and the degree of agreement of assessments both with each other and with PhenoAge (correlation > 0.89).
Among the selected 45 gut bacterial species, 16 were positively associated with age. Of these, 3 species (Muribaculum intestinale, Ruminococcus albus, Ruminococcus champanellensis) can be considered "beneficial," as they are involved in acetate production, carbohydrate fermentation, and support overall microbiota and metabolic health. However, 5 other species (Catabacter hongkongensis, Clostridium saudiense, Desulfovibrio desulfuricans, Holdemanella biformis, Howardella ureilytica) are potentially pathogenic and may cause infections or contribute to inflammatory bowel disease (IBS) involving an immune component. The remaining 8 positively associated species can be classified as neutral, as they produce acetate, butyrate, and propionate, and modulate metabolic pathways.The majority of microorganisms (29 species) exhibited a negative correlation with age, meaning their abundance decreases in older age. Among these, 7 species (Anaerobutyricum hallii, Butyricicoccus pullicaecorum, Clostridium leptum, Coprococcus comes, Eubacterium rectale, Fusicatenibacter saccharivorans, Lachnospiraceae bacterium Choco86) can be considered beneficial. They are responsible for synthesizing or fermenting various substances, support barrier function, exert anti-inflammatory effects, and reduce the risk of metabolic disorders. Conversely, only 5 species (Blautia obeum, Blautia producta, Dialister invisus, Enterocloster bolteae, Sutterella wadsworthensis) are potentially pathogenic, potentially contributing to obesity, IBS, and negatively impacting mental health. Most of the remaining age-negatively correlated species can be classified as neutral; they produce and ferment substances but under certain conditions may cause gastrointestinal disorders and metabolic disturbances.
The bacterial species used in the model collectively reflect an age-related decline in protective and metabolic functions, an increase in pro-inflammatory potential, and a disruption and impoverishment of metabolic networks.
Link: https://doi.org/10.18632/aging.206360
View the full article at FightAging
