Human beings have always wanted to know how long they will live. The arrival of inexpensive consumer genetic testing creates an obvious temptation: if ageing has a genetic component, perhaps DNA can provide the answer.
It cannot.
That does not make genetic information irrelevant to longevity. It makes the question more interesting.
The useful question is not "What age does my DNA say I will reach?" It is:
What does my genetic evidence tell me about the biological risks, pathways and characteristics that may influence the range of health and longevity outcomes I should consider?
That distinction sits at the centre of the Bioanalytix approach.
Longevity is partly heritable, but far from genetically determined
One of the seminal reviews of the genetics of human longevity was published by Kaare Christensen, Thomas Johnson and James Vaupel in Nature Reviews Genetics in 2006.
Drawing on twin studies, the authors reported that genetic differences account for approximately a quarter of the variation in adult human lifespan.
That finding is important for two reasons.
First, genes matter.
Second, most variation in lifespan is not explained simply by inherited genetic differences.
Lifestyle, environment, socioeconomic circumstances, healthcare, chance, disease exposure and interactions among these factors all contribute to the eventual outcome.
More recent genomic research has reinforced the complexity rather than replaced it with a simple genetic clock. David Melzer, Luke Pilling and Luigi Ferrucci's 2020 review in Nature Reviews Genetics describes a growing set of loci associated with lifespan and other ageing traits, including loci that overlap with age-related diseases.
The picture that emerges is not one longevity gene controlling one outcome. It is a network of biological pathways interacting with a lifetime of non-genetic influences.
Association is not destiny
Modern genome-wide association studies, or GWAS, search large populations for statistical relationships between genetic variants and traits or diseases.
This approach has transformed human genetics.
It has also created opportunities for misunderstanding.
A genetic association does not necessarily mean that a variant directly causes an outcome. Effect sizes may be small. Associations can depend on ancestry and population structure. A marker may be correlated with the causal biology rather than itself being causal. Different environmental exposures can alter the significance of inherited susceptibility.
Tam and colleagues' review in Nature Reviews Genetics describes both the considerable achievements of GWAS and these important limitations.
For Bioanalytix, the implication is straightforward:
a genetic observation should be treated as evidence with a strength and context, not as a deterministic instruction.
APOE shows why the distinction matters
Few examples illustrate this better than APOE.
Variation in APOE is associated with Alzheimer's disease susceptibility and longevity. The ε4 allele is associated with increased Alzheimer's disease risk, while other APOE configurations have different risk profiles.
But carrying APOE ε4 does not mean that an individual will develop Alzheimer's disease.
GeneReviews explicitly notes that the presence of one or two ε4 alleles increases risk but does not constitute a diagnosis. Some carriers never develop the disease, while Alzheimer's disease also occurs in people without ε4.
The information is therefore potentially meaningful without being deterministic.
That distinction matters enormously.
A responsible interpretation is:
this result may change the evidence about susceptibility.
An irresponsible interpretation is:
this result tells us what will happen to you.
Bioanalytix is designed around the first interpretation.
FOXO3 illustrates another kind of longevity evidence
FOXO3 provides a different example.
Variants in FOXO3 have shown replicated associations with longevity across multiple human populations. Research has connected the FOXO3 pathway with biological processes including energy metabolism, oxidative stress, proteostasis, inflammation, cellular maintenance and stress responses.
This makes FOXO3 scientifically interesting.
It does not make it a longevity guarantee.
An individual carrying a longevity-associated FOXO3 variant cannot infer a particular lifespan from that observation. At most, the result contributes another piece of evidence to a much larger biological picture.
This is characteristic of complex traits.
The value often lies not in one marker but in understanding how multiple pieces of evidence alter what we know — and what remains uncertain.
Disease susceptibility and longevity are connected
Longevity cannot be considered independently of health.
Cardiovascular disease, metabolic disease, neurodegenerative disease, cancer and other age-related conditions influence both lifespan and healthspan. Genetic variants associated with these conditions can therefore be relevant to a broader longevity assessment even when they are not conventionally described as "longevity genes."
This produces an important distinction between lifespan and healthspan.
Lifespan asks how long someone lives.
Healthspan asks how much of that life is lived in relatively good health and functional independence.
For long-term planning, both matter.
A long life with decades of independence presents a different financial problem from a long life involving substantial care needs. A shorter-than-assumed planning horizon presents another problem again.
DNA cannot tell us which future will occur.
It may, however, provide evidence that helps us think more intelligently about the alternatives.
The evidence itself evolves
Genetic science is not static.
Large biobanks, larger GWAS datasets, improved statistical methods, functional genomics and more diverse study populations continue to reveal new associations and challenge older ones.
This means that a genetic observation should not be regarded as permanently interpreted.
Bioanalytix therefore distinguishes between different levels of evidence.
Established evidence has replicated scientific support sufficiently strong to justify meaningful attention.
Emerging evidence has credible support but remains dependent on context, replication or developing science.
Exploratory evidence may be scientifically interesting but should not drive a conclusion by itself.
These categories are intentionally conservative.
A system dealing with genetic evidence should be able to say not only what an association is, but also how much confidence should reasonably be placed in it.
Why Bioanalytix does not predict a death age
Suppose a model took several genetic markers, combined them with population life tables and produced the statement:
"Your predicted lifespan is 91."
The precision would be appealing.
It would also conceal enormous uncertainty.
The estimate would depend on model specification, population assumptions, environmental factors, future medical developments, behaviour and many events that cannot be observed today.
Adding decimal places would not solve that problem.
Bioanalytix therefore takes a different approach.
The objective is not to collapse uncertainty into a single predicted age. It is to identify evidence that may make particular biological and longevity scenarios more or less worthy of attention.
That is a much more modest claim.
It is also potentially much more useful.
From biological evidence to planning uncertainty
The connection to financial planning follows from a simple observation.
Retirement decisions depend on an uncertain horizon.
If someone retires at 65, there is an enormous financial difference between needing a portfolio to support 15 years of retirement and needing it to support 35 or 40 years.
Yet the correct response is not to replace a population-average life expectancy with a genetically predicted death date.
Instead, biological evidence can contribute to the assessment of which longevity and health scenarios should be tested.
Those scenarios can then be used to explore questions such as:
- How long might retirement assets need to provide income?
- How sensitive is the plan to a longer retirement?
- How much liquidity might be required?
- How does the appropriate balance between growth and defensive assets change across scenarios?
- What happens to sustainable withdrawals?
- What happens to estate capacity?
- Which assumptions matter most to the result?
This is where biological information can acquire economic relevance without pretending to possess predictive certainty.
The important output is not certainty
The deepest misconception about personalised genetics is that its value must come from prediction.
Sometimes the opposite is true.
Genetic evidence can be useful because it helps us understand uncertainty more intelligently.
For Bioanalytix, that is the objective.
Your DNA is not a timetable for your life.
It is one source of evidence about the biological system in which your financial future will unfold.
And the planning question is not:
"When will I die?"
It is:
"What futures should I be prepared to finance?"
References
Christensen, K., Johnson, T. E. & Vaupel, J. W. (2006). The quest for genetic determinants of human longevity: challenges and insights. Nature Reviews Genetics, 7, 436–448. https://doi.org/10.1038/nrg1871
Melzer, D., Pilling, L. C. & Ferrucci, L. (2020). The genetics of human ageing. Nature Reviews Genetics, 21, 88–101. https://doi.org/10.1038/s41576-019-0183-6
Tam, V., Patel, N., Turcotte, M., Bossé, Y., Paré, G. & Meyre, D. (2019). Benefits and limitations of genome-wide association studies. Nature Reviews Genetics, 20, 467–484. https://doi.org/10.1038/s41576-019-0127-1
Morris, B. J., Willcox, D. C., Donlon, T. A. & Willcox, B. J. (2015). FOXO3: A major gene for human longevity — a mini-review. Gerontology, 61(6), 515–525. https://doi.org/10.1159/000375235
Bird, T. D. Alzheimer Disease Overview. GeneReviews. National Center for Biotechnology Information. https://www.ncbi.nlm.nih.gov/books/NBK1161/