What a Genome Cannot Tell You

Message us and we will tell you if this would not help you, or call 020 3951 3429.

When to get help today

A genome answers nothing urgent. If you have chest pain, breathlessness at rest, or you lose consciousness without warning, call 999 or go to your nearest emergency department today.

If you have a symptom now, get the symptom diagnosed now. Sequencing is a poor way to work out what is wrong today and a good way to delay finding out.

The things it cannot tell you

It cannot tell you how long you will live. Nothing in a sequence carries that, and no provider who claims otherwise can show you the study that supports the claim. The list below is what a genome is silent about, in the order people are most often surprised by.

It cannot rule out the illnesses most people get. Most heart disease, most cancer, most diabetes and almost all infection arise from a mixture of inheritance, environment, behaviour and chance, and the inherited part is spread thinly across thousands of positions rather than concentrated in one readable gene. A clear report on the ACMG list means no finding on that list. It does not mean a clear future.

It cannot read every part of the genome equally well. Long repetitive stretches and large structural rearrangements are harder to call from short read sequencing than single letter changes are, so a normal result is a normal result for the classes of variant the method detects.

It cannot see what your cells do later. Most cancers arise from changes acquired in a tissue over a lifetime, and those are not in the genome you were born with.

A variant of uncertain significance is not an answer

A large share of what sequencing finds sits in a middle category. The laboratory can see that a letter differs from the reference and cannot say whether it matters.

That is neither a diagnosis nor an all clear, and the correct response to one is usually to do nothing and let it be re read as evidence accumulates. It is also the single commonest way people are harmed by this. A person told they carry an uncertain variant buys scans, changes plans and worries for years about something that is later reclassified as harmless.

A clinic that reports uncertain variants to you without a counsellor in the room is transferring a problem to you rather than solving one. Ask any provider what they do with this category before you buy.

Message us about a variant you have already been told about

A finding is a probability, not a sentence

The ACMG secondary findings list at version 3.3, published in Genetics in Medicine in 2025, names 81 genes chosen because a finding changes what a doctor does. It is a list built on actionability rather than on certainty.

Around 6% of people carry a finding on it that changes management, on a cohort of 3,972 exomes read against version 3.3 of that list, Genetics in Medicine, 2025, of which about 3.0% were cardiovascular and about 2.0% cancer related. Carrying such a variant raises risk. It does not settle the question of whether the condition will appear, and how often it appears differs by gene and is generally lower in people found by screening than in families found through a relative who was already ill.

That distinction is what the counsellor spends the appointment on, and it is why a result is delivered by a person rather than a portal.

What polygenic risk does and does not do

The European Society of Cardiology published a clinical consensus statement on polygenic risk scores in the European Heart Journal in 2025. The position it supports is narrow and worth reading precisely.

A polygenic score is most useful in the middle of the risk range, where a person sits close to a treatment threshold and the score moves the decision one way or the other. At the extremes it usually changes nothing, because the decision was already made by age, blood pressure, cholesterol and smoking.

The other limit is ancestry. Most of the data these scores were built from came from people of European ancestry, and a score transported to a different population performs less well. Any provider selling you a polygenic number without telling you which population it was derived in is selling you a number with an unstated error bar.

Where the medicines evidence stops

PREPARE is the best trial there is and its absolute effect is modest. Swen and colleagues, in The Lancet in 2023, found clinically relevant adverse drug reactions in 21.5% of the genotype guided group against 28.6% of controls across 6,944 patients, an odds ratio of 0.70. That is about 7 percentage points in absolute terms, the trial was open label, randomisation was clustered by country and site rather than by patient, and a published correspondence in the same journal questioned whether an effect of that size justifies testing whole populations.

TAILOR-PCI, testing genotype guided antiplatelet choice after a stent, did not reach significance. Pereira and colleagues, in JAMA in 2020, reported 4.0% against 5.9% at 12 months, a hazard ratio of 0.66 with a 95% confidence interval of 0.43 to 1.02 and p equal to 0.06.

And where the stakes are highest the correction is incomplete. Henricks and colleagues, in Lancet Oncology in 2018, found severe toxicity in 39% of DPYD carriers against 23% of non carriers across 1,103 evaluable patients. The 25% dose reduction used for two variants did not bring risk back to normal. Only the 50% reductions came close.

Who should not buy this

A healthy 30 year old with no family history, no medicines and no plans to change anything is buying reassurance rather than information. Reassurance is a legitimate purchase and it should be made knowingly.

Anyone with a symptom right now should get the symptom diagnosed first.

Anyone whose family already has a known genetic variant should have targeted testing for that specific variant, usually through the clinical genetics service already looking after the family. It is faster, it is cheaper and it gives a cleaner answer than sequencing everything.

Anyone who would not act on a finding. If you would not change a screening interval, take a medicine, or tell a relative, the file will change nothing.

Anyone buying it for a child without a clinical reason. We do not sequence children for adult onset conditions on request. The decision belongs to the adult that child becomes.

Anyone for whom an uncertain result would be intolerable. Say so in the first conversation and we will tell you honestly whether to proceed.

What you can do instead, for nothing

Four questions answer most of this before any money changes hands, and they are better answered on paper than in a consulting room. Write them down and take the answers to whoever you speak to next, here or anywhere else.

What would I do differently if the answer were bad. What would I do differently if the answer were normal. Who else in my family does this result belong to. And would I still want to know in five years if nothing could be done about it.

Then do the two things that are free and better evidenced than anything on this page. Write down a three generation family history with ages and causes, because anything before 50 carries far more weight than anything after 70. And get your blood pressure, cholesterol and blood sugar measured, because those three change what happens to most people far more than any sequence does.

What we would do

One. A conversation with a registered genetic counsellor, at no charge, in which we may tell you not to proceed at all. Nothing is taken and nothing is committed on that call, and a fair share of them end in a recommendation to wait.

Two. If you do proceed, saliva at home or a blood draw in Marylebone, sequenced at 30x in a laboratory accredited to ISO 15189.

Three. The result read with a doctor, written into your record, and re read every year, including the reclassification of anything uncertain.

Genome is £4,995 and it includes the counsellor before and after, the doctor review and three yearly re reads. The first conversation is not charged whether or not you proceed.

Start with the first conversation

Message us to talk it through first

Questions people ask

Can whole genome sequencing predict how long I will live? No. Nothing in a sequence carries that. It reads a defined set of genes for findings that change management and is silent about the mixture of environment, behaviour and chance that determines most outcomes.

If my genome is clear, am I safe? No. A clear report means no finding on the list it was read against. Most heart disease, cancer and diabetes are not caused by a single readable gene, and short read sequencing calls some classes of variant less well than others.

What is a variant of uncertain significance? A letter that differs from the reference where the laboratory cannot say whether it matters. It is neither a diagnosis nor an all clear. The usual correct response is to do nothing and let it be re read as evidence accumulates, which is why the result is delivered by a counsellor.

Are polygenic risk scores reliable? They are most useful in the middle of the risk range, where they move a decision, according to the European Society of Cardiology consensus statement in the European Heart Journal in 2025. They perform less well outside the mainly European ancestry populations they were derived in.

Should I have this if someone in my family has a known genetic condition? Usually not first. Targeted testing for that specific variant, through the service already looking after your family, is faster, cheaper and cleaner. We will say so and point you there rather than sell you a genome.

Will you tell me not to buy it? Yes, when that is the answer. The first conversation with the counsellor is not charged and does not commit you to a sample, and it exists partly so that people who should not proceed do not.

Can I have my child sequenced? Not for adult onset conditions on request. Where there is a clinical reason we will arrange it properly, and where there is not, the decision belongs to the adult that child becomes.

Does a normal result mean I can stop screening? No. Age based screening exists because most disease is not explained by a single gene, and a normal genome is not a reason to come out of any programme you are eligible for.

Message us before you spend anything, or call 020 3951 3429.

This page is information and not a diagnosis. Sequencing does not replace examination, history or blood work, and no result guarantees an outcome. If you are unwell today, see a doctor today. In an emergency call 999.

References

Krogsbøll LT, et al. General health checks in adults for reducing morbidity and mortality from disease. Cochrane Database of Systematic Reviews, 2019. 10.1002/14651858.CD009009.pub3

Swen JJ, van der Wouden CH, Manson LE, et al. A 12-gene pharmacogenetic panel to prevent adverse drug reactions. The Lancet, 2023;401(10374):347-356. 10.1016/S0140-6736(22)01841-4

Pereira NL, Farkouh ME, So D, et al. JAMA, 2020;324(8):761-771. PMID 32840598.

Henricks LM, Lunenburg CATC, de Man FM, et al. Lancet Oncology, 2018;19(11):1459-1467.

American College of Medical Genetics and Genomics. Recommendations for reporting of secondary findings in clinical exome and genome sequencing, ACMG SF v3.3. Genetics in Medicine, 2025.

European Society of Cardiology. Clinical consensus statement on polygenic risk scores in cardiovascular disease. European Heart Journal, 2025.

Van Driest SL, et al. Clinical Pharmacology and Therapeutics, 2014;95(4):423-431.

Medicines and Healthcare products Regulatory Agency. Drug Safety Update, October 2020, DPD deficiency and fluoropyrimidines.

ISO 15189:2022. Medical laboratories, requirements for quality and competence. International Organization for Standardization.

Clinical Pharmacogenetics Implementation Consortium. Published gene and drug guidelines, cpicpgx.org, read 8 September 2026.