A concept, method, algorithm, and computer system (CS) of step-by-step dialog optimization of biomarker (BM) panels for assessing human biological age (BA) according to a number of universal criteria based on incomplete and noisy data have been developed. This system provides the ability to automatically build BM panels for BA assessment and to increase the accuracy of BA determination while reducing the number of measured BMs. The optimization criteria are as follows: high correlation of BMs with chronological age (CA); minimum size of BM panels, obtained by rejecting highly cross-correlated BMs; high accuracy of BA assessment; high accuracy of BA/CA dependency interpolation; absence of outliers in BM values, which reduce the BA assessment accuracy; rejection of panels resulting in a high standard deviation for the BA-CA difference; and possible additional criteria entered by the researcher according to the task specifics. The CS input consists of data on physiological, biochemical, and other BMs that change with age. The CS output is a panel of BMs optimized according to the specified optimization criteria. The CS is user-friendly. It allows the user to add optimization criteria that the researcher considers to be important or to remove criteria that the user considers incorrect. The CS may be used in solving practical problems of anti-aging medicine, such as the treatment and prevention of age-related chronic non-infectious diseases representing the main causes of death. The authors' point of view on the role and place of BA diagnostics in this area is discussed.
Keywords: aging; biological age; biomarkers of aging; diagnostics of aging; dialog optimization; personalized medicine.
Copyright © 2021 Krut'ko, Dontsov, Ermakova, Makarova, Mitrokhin, Shashina and Shcherbakov.