Biology and complex systems thinking are at the core of Oken’s algorithms.
All living organisms keep the many molecules that compose their bodies in a state of equilibrium. This dynamic process is called homeostasis and is vital for maintaining good physical, mental and biological health with changing environmental and physiological conditions.
Complex systems are dynamic systems made of different components/agents that can interact with one another in intricate ways, leading to nonlinear and emergent characteristics not captured by the sum of its parts. Organisms are perfect examples of such complex systems at many levels: molecules interact in large signaling networks to keep the cells in the right functional states; cells send messenger molecules to other cells in coordinated fashion to maintain tissue and organ integrity; immune cells cooperate to provide defense against foreign agents.
When physiological systems lose homeostasis (‘’dysregulation’’), with aging or diseases, levels of many molecules in the systems will change in a coordinated manner. These changes can be hard to detect by measuring single molecule, but can more reliably be detected by analysing many molecules from the system in parallel.
Our approach uses simple blood biomarker levels, such as those regularly measured by physicians, ex. cholesterol, hemoglobin, vitamins, calcium, and albumin, to assess the loss in homeostasis. The algorithms integrate the biomarker values to quantify how abnormal an individual’s biomarker profile is in comparison to the reference population’s norm. This idea was developed by Pr Alan Cohen and his research team at the University of Sherbrooke in Canada, and has undergone extensive optimization to increase predicting power. Indeed, one individual can have all biomarker levels in clinically acceptable ranges but still have an abnormal profile based on our algorithms. Higher scores can predict mortality, frailty, physical function decline, cognitive decline, depression and multiple chronic age-associated diseases.
Our algorithms can calculate your global health score to give you an overall idea of your health state. We can also generate system-specific scores to give more precise insights into specific biological systems functioning, such as the immune system, liver and kidney functions, oxygen transport system or vitamins. We know from Pr Cohen’s research that scores for each systems give you different pieces of information on your physiological functioning, the scores being weakly correlated to each other. In addition, all systems have different predicting power for different age-associated changes.
As with any tool, our algorithms have certain limitations.
Why Oken’s algorithms are not biological age?
While they do correlate with age, the scores calculated by our algorithms are much better predictor of health status and age-related outcomes than of age per se.
Oken’s algorithms are not diagnosis or prognosis
Dysregulation is a complex and multi-dimensional process. No single metric can adequately summarize all of the dysregulation happening at a given moment. We believe DSign is a reasonable compromise between cost, feasibility, generalizability, and information content, but it is by all mean NOT a validated diagnostic or prognostic tool.
Not all high scores are bad
Studies have largely been conducted at the population level, where we have detected substantial heterogeneity in DSign across individuals. In other words, some individuals have consistently higher or lower DSign scores than others. Because each individual has a unique genetic and environmental context, it is not possible to say that every individual with a higher DSign score is in worse health, though this tends to be the case in a population. Within an individual, it is likely that a consistent, longterm increase in DSign would indicate a health problem, but it would be dangerous to overinterpret a single DSign score for an individual out of context without further validation.