APOE and Alzheimer's: Reading Two SNPs in Your Raw Data
Your APOE type is set by rs429358 and rs7412. How to read them, what e4 does and does not mean, why some services hide it, and how to decide whether to look.
APOE is the best-known gene in Alzheimer’s research and, for many people, the single most emotionally charged thing in a raw DNA file. It is also unusually easy to read: your APOE type comes from just two SNPs, both of which sit in most consumer files. Before you go looking, it is worth understanding what the answer would and would not tell you.
What APOE does
Apolipoprotein E is a protein that ferries cholesterol and other fats around the body and the brain. Humans carry three common versions, named e2, e3 and e4, which differ at two amino-acid positions. e3 is the most common. e4 is the version associated with higher Alzheimer’s risk, and e2 with somewhat lower risk. Everyone carries two copies, so your APOE type is a pair: e3/e3 for most people, e3/e4 for roughly a quarter of people of European ancestry, e4/e4 for about two percent.
The two SNPs that define it
The two positions map to two SNPs on chromosome 19, and raw files report them on the forward strand:
- rs429358 - the e4 position. The common allele is T; C is the e4 allele.
- rs7412 - the e2 position. The common allele is C; T is the e2 allele.
Read together:
| rs429358 | rs7412 | APOE type |
|---|---|---|
| TT | CC | e3/e3 |
| CT | CC | e3/e4 |
| CC | CC | e4/e4 |
| TT | CT | e2/e3 |
| TT | TT | e2/e2 |
| CT | CT | e2/e4 (almost always) |
The last row is the one place the two SNPs are ambiguous. Two heterozygous results could in principle be e2/e4 or e1/e3, but e1 is vanishingly rare, so e2/e4 is the safe reading.
Both markers are on current 23andMe and AncestryDNA chips. Some older chip versions lacked rs429358 or reported it under an internal identifier, so if you cannot find it, that is a chip limitation rather than a hidden result.
What e4 actually means
The landmark finding came from Corder and colleagues in 1993: the risk of late-onset Alzheimer’s rose with each e4 copy carried. In people of European ancestry, one copy raises lifetime risk roughly two to three times relative to e3/e3, and two copies raise it far more - often quoted around eight to twelve times, though the exact figure depends heavily on the population and the study. The effect is weaker in people of African ancestry and in some Hispanic populations, so a risk number from a European cohort does not transfer cleanly.
Three things keep this in proportion:
- e4 is neither necessary nor sufficient. Many e4 carriers never develop the disease, and a large share of people with Alzheimer’s carry no e4 at all.
- It shifts age of onset as much as whether. e4 carriers who develop Alzheimer’s tend to do so earlier, which is part of why the lifetime risk numbers look large.
- Most of the risk is still not APOE. Age, cardiovascular health, education, hearing loss and dozens of smaller genetic variants all contribute.
A 2024 paper in Nature Medicine argued that e4/e4 should be considered a distinct genetic form of the disease rather than merely a risk factor, based on how consistently homozygotes showed early biological changes. That framing is debated, but it illustrates why the two-copy case is treated differently from one copy.
Why services make you opt in
23andMe and others put APOE behind a separate consent step for good reason. Learning you carry e4 can cause real distress without offering a clear action, because there is no proven prevention that is specific to e4 carriers. The interventions that help - blood pressure control, exercise, sleep, hearing correction, not smoking - are the same ones recommended to everyone.
There are also practical consequences. In many places, genetic information can affect long-term care, life and disability insurance underwriting even where health insurance is protected; our piece on DNA and life insurance explains the gap. And once you know, you cannot un-know, which matters for relatives who share your APOE type at 50 percent.
Deciding whether to look
Some people want to know so they can plan, join prevention trials, or simply satisfy curiosity. Others would rather not carry a number around for decades. Both are reasonable. A few suggestions if you are on the fence:
- Decide before you search, not after the letters are on screen.
- Consider talking to a genetic counsellor first, especially if Alzheimer’s runs in your family.
- If you do look, do it with a tool that keeps the file on your own device rather than uploading it somewhere that stores it.
If you decide to proceed, the two rsIDs above are all you need, and a plain text search of your file will show them. The Genespiral marker pages describe what each genotype is associated with in the same terms used here.
References
- Corder EH, et al. Gene dose of apolipoprotein E type 4 allele and the risk of Alzheimer’s disease in late onset families. Science. 1993. PubMed 8346443
- Fortea J, et al. APOE4 homozygozity represents a distinct genetic form of Alzheimer’s disease. Nature Medicine. 2024. PubMed 38710950
- National Institute on Aging. Alzheimer’s Disease Genetics Fact Sheet.
- dbSNP entries for rs429358 and rs7412, NCBI.
This article is educational only and is not medical advice. If you are worried about dementia risk, speak with a doctor or a genetic counsellor.