Multi-omic models
SystemAge
How to interpret a proprietary, system-labeled aging estimate without mistaking the model for a direct organ assessment.
What it is
A model built from several kinds of signal
SystemAge refers to a proprietary approach that combines biological inputs and reports modeled estimates across named systems or pathways. This is a multi-omic interpretation layer: algorithms look for patterns in laboratory data and compare them with patterns in a reference population.
These modeled estimates are not direct measurements of organ function or disease. A heart-labeled or liver-labeled output is not equivalent to an ECG, imaging study, liver chemistry, physical examination, or diagnosis.
The output can be useful for understanding how a model organizes data, but clinical utility and treatment guidance remain uncertain. There is not yet a validated rule that turns each system estimate into a therapy or predicts what will happen to an individual.
Under the report
Four layers shape the estimate
Biological inputs
A multi-omic approach may combine epigenomic, proteomic, metabolomic, or other laboratory signals. The exact inputs and weighting depend on the assay and model.
Reference data
The model learns relationships from a particular population. Performance may not transfer equally across ages, ancestries, health states, or collection settings.
Modeled outputs
A report can summarize patterns under organ-system or pathway labels, but the label remains an estimate produced by the model rather than a direct functional test.
Version and platform
Laboratory methods, normalization, and model versions can change. Those differences matter when comparing a later result with an earlier one.
Interpreting a result
Treat the label as a prompt, not a conclusion
If a system estimate seems surprising, begin with history, symptoms, established risk factors, and validated clinical tools. A modeled signal may generate a question, but it cannot confirm the affected tissue, explain causation, or establish severity.
Repeating the assay is most interpretable when collection, laboratory processing, and model version are comparable. Even then, the amount of change that exceeds analytical and biological variability may be unclear.
Questions for any proprietary model
- What exactly was measured, and what was inferred?
- What population and outcomes were used to train and validate the model?
- How reproducible is the result within the same person?
- What independent evidence shows that acting on the output improves health?
Put this model in context
Read the broader biological-age overview, compare it with an IgG-glycan-derived estimate, or review how to choose standard laboratory tests.