There are many ways in which biological data can be processed via machine learning techniques to produce clocks that assess the burden of aging. Today's open access paper draws a distinction between aging clocks, which provide information on biological age, and mortality timers, which provide information on risk of death. Aging affects everything in the body, and all aspects of physiology and cellular biochemistry undergo at least some change. So epigenetic marks on the genome, levels of transcription of various genes, circulating proteins in the bloodstream, the pattern of microbial populations in the gut microbiome, specific chemical modifications to proteins, and much more can be assessed in bulk and then mined for associations with age and mortality risk. Even physical measures such as grip strength, ability to stand from sitting, walking speed, and so forth, can be algorithmically combined to form clocks.
With so much data to use in the production of clocks, it is inevitable that the quality and applicability of individual clocks will vary widely on a case by case basis. It is at present always unclear as to how the specific metrics that form the clock are caused by specific underlying processes of aging. That isn't an obstacle to the use of the clock in studies of natural aging, but it is a roadblock to the use of a clock as a way to assess the success of a potential age-slowing or age-reversing interventions. Interventions that address mechanisms of aging only impact a limited set of those mechanisms, or just one mechanism. A clock may be overly weighted towards that mechanism, or it may be insensitive to that mechanism, and in either case the results will be misleading. A great deal of work has yet to be accomplished to allow clocks to reach their true potential, as ways to steer medical development towards the most effective approaches to rejuvenation.
Aging clocks can be devised from any biological system that changes during age. Measuring the amount of variation in those biological systems may allow scientists to peer into how far an organism has drifted from youthful function or how close they are to mortality. Aging clocks specifically aim to inform subjects of their biological age, but many of these clocks deliver no information on how long a subject may have left. However, in cases where the data can deliver information on impending death (unless some type of intervention is taken), then those clocks are also mortality timers that can better serve a subject's decision making or the advice from a healthcare practitioner. Three categories exist, aging clocks that deliver biological age, aging clocks and mortality timers that deliver biological age and information to predict death, and mortality timers that only offer information that can be used to predict the onset of disease or death.
Even though aging clocks have become popular, the term mortality timer should also be used (where applicable) across the industry to deliver sobering information that may assist in changing poor decision making related to a subject's lifestyle. Many aging clocks and mortality timers exist, such as blood biomarkers (proteomics), epigenetic mechanisms, extracellular vesicles, immune system factors, telomere length, glycomic levels, grip strength, blood vessel health, and many more biological systems could be used to determine a subject's age. However, the accuracy to predict age, overall health, or impending demise may be lacking. From analyzing various aging clocks to mortality timers, it becomes evident that even though some aging clocks may deliver vast amounts of biological data, no single clock can deliver all biological data across all tissues from the generic samples collected with commercial kits.
All clocks appear to have limitations on the data they can deliver, albeit the data are extremely extensive from some aging clocks such as epigenetic, blood tests, and glycomic clocks. The findings of this article are (in no order of effectiveness) that epigenetic, glycomic, and blood/serum biomarkers are the three most powerful clocks that can be used, as not only can a biological age prediction be made, along with disease potential to disease penetrance, but they can also function as rough mortality timers. An impending mortality diagnosis is a sobering prediction to any subject, so clocks that can also deliver rough time of death also function as extreme motivation for subjects to change their lifestyle habits as a matter of urgency. Aging clocks will continue to evolve as new biomarkers are found; however, any biological machinery that wanes with age can be used to elucidate data regarding biological age.