Alcohol and your DNA: ADH1B and ALDH2

Genetics explainer · Updated June 2026

The two-step pathway

Alcohol leaves your body through a short relay of two enzymes, and understanding those two steps explains almost everything about why people react so differently to the same drink:

1. Ethanol → acetaldehyde, handled mainly by alcohol dehydrogenase (the ADH enzymes, including the one made by the ADH1B gene).
2. Acetaldehyde → acetate (harmless), handled by aldehyde dehydrogenase 2 (ALDH2). Acetate is then broken down further into water and carbon dioxide and cleared from the body.

Acetaldehyde is the troublemaker in the middle. It's chemically reactive, toxic to cells, and recognized as a carcinogen, and it's the molecule that causes facial flushing, a pounding or racing heart, headache, and nausea when it builds up. The ethanol itself makes you feel intoxicated, but it's the acetaldehyde that tends to make you feel sick.

The crucial point is that acetaldehyde is only an intermediate. In most people it exists for a very short window before ALDH2 converts it to harmless acetate, so it never accumulates enough to cause symptoms. Whether it piles up instead comes down to the balance between the two steps: how fast step one produces it versus how fast step two clears it. Speed up production, slow down clearance, or both, and the intermediate spends longer in your bloodstream and tissues. That single idea — a fast-filling tank and a slow drain — is the mechanism behind everything else in this post.

ADH1B: how fast you make acetaldehyde

The ADH1B gene comes in variants that differ dramatically in speed. The well-studied variant rs1229984 (His48Arg) produces a "fast" form of the enzyme:

• The fast ADH1B allele converts alcohol to acetaldehyde much more quickly than the common form.
• That means acetaldehyde can spike soon after drinking — even in people whose ALDH2 clearance is normal.

On its own, a fast ADH1B tends to make drinking less comfortable: the unpleasant acetaldehyde phase arrives faster and hits harder. Even when the ALDH2 clearance enzyme is fully working, a rapid first step can briefly outpace it, so acetaldehyde surges before it's mopped up.

The fast allele is most common in people of East Asian descent but also appears at meaningful frequencies in some Middle Eastern, European, and other populations — which is one reason this marker is worth understanding regardless of ancestry. It's a good example of why alcohol response isn't captured by a single "flush gene": the same visible reaction can trace back to different points in the pathway depending on your genotype.

ALDH2: how fast you clear it

If ADH1B is the accelerator on the first step, ALDH2 is the brakes on the second. The rs671 variant (Glu504Lys) largely inactivates the ALDH2 enzyme:

• Active ALDH2 clears acetaldehyde efficiently.
• Inactive ALDH2 lets acetaldehyde pile up — producing the strong, visible flush reaction.

This is the variant behind the so-called "Asian glow." Importantly, rs671 behaves in a partly dominant way: carrying a single copy of the inactive allele already leaves you with substantially reduced acetaldehyde-clearing capacity, and carrying two copies leaves you with very little. That's why some people flush after a few sips while others can drink more before it shows — the difference often comes down to whether they carry one inactive copy or two.

Because ALDH2 is the more familiar and dramatic of the two markers, we cover it in depth — including the full genotype table and step-by-step instructions for finding rs671 in your raw data — in the dedicated ALDH2 flush guide.

Why the two genes interact

The reason these markers are usually discussed together is that the effect depends on both steps at once:

• Fast ADH1B + inactive ALDH2 is the most uncomfortable combination — acetaldehyde is made quickly and cleared slowly, so it builds up fast. People with this pairing often feel unwell after very little alcohol.
• Fast ADH1B + active ALDH2 still makes acetaldehyde quickly, but clears it, so the discomfort is usually milder.
• Common ADH1B + active ALDH2 is the typical "no strong reaction" profile.

This is why two people who both flush can have quite different underlying genetics — and why the flush isn't a single on/off switch. It's a continuum set by two independent dials. The fast ADH1B dial controls how quickly the tank fills; the ALDH2 dial controls how quickly it drains. The worst-case acetaldehyde exposure comes from turning both dials against you at once: making it quickly and clearing it slowly.

Because the two genes sit on different chromosomes and are inherited independently, your combination is essentially a roll of two separate dice. That also means a "reassuring" result at one gene doesn't cancel a concerning result at the other — they describe different steps, and both matter.

Why these variants are common in certain ancestries

It might seem strange that variants which make drinking unpleasant became common at all. The leading hypothesis is that they spread through populations for reasons that had little to do with drinking as we think of it today. The fast ADH1B and inactive ALDH2 alleles rose to high frequency in parts of East Asia over the last several thousand years, and one prominent idea links this to the spread of rice cultivation and fermentation: in settings where fermented beverages and contaminated water were part of daily life, a genotype that made heavy drinking physically aversive — or that changed how ethanol and acetaldehyde were handled — may have carried some survival advantage. The full story is still debated, and you should treat any tidy adaptive narrative with caution.

What's clear is the population pattern itself: both variants are found at their highest frequencies in East Asian populations, the fast ADH1B allele also appears in some Middle Eastern and European groups, and the inactive ALDH2 allele is comparatively rare outside East Asia. This is why ancestry context helps you interpret a result, but it never replaces reading your own genotype — plenty of individuals carry these variants against the "expected" background of their ancestry, and plenty don't.

What it's associated with

A few associations are well established in the research literature:

• Lower alcohol dependence. Both the fast ADH1B allele and the inactive ALDH2 allele are consistently linked to lower rates of alcohol dependence. The leading explanation is straightforward: when drinking produces acetaldehyde discomfort sooner, people tend to drink less, and a substance that reliably makes you feel unwell is harder to build a habit around. This is one of the better-replicated gene–behavior associations in all of human genetics precisely because the mechanism is so direct.
• Cancer risk among those who do drink. Because acetaldehyde is a recognized carcinogen, people who carry the inactive ALDH2 variant and continue to drink regularly have an elevated risk of cancers of the esophagus and the head and neck — the upper-GI tissues that the intermediate contacts first. The key nuance is that this risk is tied to alcohol intake, not to carrying the gene by itself. Someone with the inactive variant who drinks little or not at all is not exposing those tissues to sustained acetaldehyde. The variant doesn't cause cancer; it changes how much acetaldehyde a given amount of drinking produces, which is why it functions as an evidence-based reason for moderation rather than a diagnosis.

Framed plainly: if you know you carry the inactive ALDH2 variant, the useful takeaway isn't alarm — it's that "moderate" may mean something different for you than for someone who clears acetaldehyde quickly, and that this is a genuinely worthwhile thing to weigh when you decide how much to drink.

Heavy drinking, whatever your genotype, also tends to show up on routine bloodwork — it's a common reason for elevated liver enzymes, where an AST higher than ALT is the classic alcohol-related pattern.

What a positive result should — and shouldn't — mean

Say you check your raw data and find you carry the fast ADH1B variant, the inactive ALDH2 variant, or both. What should you actually do with that?

What it does tell you: it explains a real, mechanistic part of how your body handles alcohol — the speed of each step in the pathway — and it gives you a concrete, research-backed reason to think about moderation if you drink. If you've always wondered why you flush, feel sick fast, or never developed much of a taste for alcohol, this may be a large part of the answer.

What it shouldn't mean: it isn't a verdict. A variant here doesn't tell you that you will develop a health problem, and a reassuring "common ADH1B, active ALDH2" result doesn't mean alcohol is harmless for you. These two genes describe two enzyme steps; they don't capture dose, frequency, liver health, other genes, medications, or the many other factors that shape alcohol's effects. The result is a starting point for a better-informed conversation, not the end of one.

Why raw data isn't diagnostic

It's worth being precise about what a 23andMe or AncestryDNA export actually is. These are genotyping arrays: they read specific, pre-chosen positions in your genome — including well-studied ones like rs1229984 and rs671 — but they are consumer research tools, not clinical diagnostic tests. Individual positions can occasionally be miscalled, arrays don't cover everything, and a genotype is a probability-shifting piece of information, not a measurement of your body's current state.

Quanome reads these markers from your raw file entirely on your device, so you can explore what you carry without uploading your genome anywhere. But exploring is the right word: the honest use of this data is to become better informed and to bring good questions to someone qualified to answer them, not to self-diagnose.

For genotype-by-genotype detail, browse the Quanome gene library. For the flush reaction specifically and how to find rs671 in your raw data, see the ALDH2 flush guide. Or explore the rest of the Quanome blog.

Frequently asked questions

What's the difference between ADH1B and ALDH2?

They work on different steps. ADH1B controls how fast alcohol is converted into acetaldehyde — a fast variant makes that first step happen quickly. ALDH2 controls how fast acetaldehyde is then cleared away into harmless acetate. A problem at either step lets acetaldehyde build up, which causes flushing and nausea.

Does ADH1B cause the alcohol flush?

Indirectly. A fast ADH1B variant produces acetaldehyde more quickly, so it can accumulate faster — contributing to flushing and feeling unwell. The classic, dramatic flush reaction is most strongly driven by the ALDH2 rs671 variant, which slows acetaldehyde clearance.

Why are these variants linked to lower alcohol dependence?

Both the fast ADH1B and the inactive ALDH2 variants make drinking less pleasant — alcohol produces acetaldehyde-related discomfort sooner. People who feel sick faster tend to drink less, and population studies consistently associate these variants with lower rates of alcohol dependence.

Where do these variants come from?

The fast ADH1B (rs1229984) and inactive ALDH2 (rs671) variants are both most common in people of East Asian descent, though the fast ADH1B allele also appears at notable frequencies in some other populations. This is educational, not medical advice.

Can 23andMe raw data diagnose my alcohol tolerance?

No. Raw DNA from 23andMe or AncestryDNA is not a diagnostic test. It can show which version of rs1229984 and rs671 you carry, which is genuinely informative about the enzyme pathway, but it doesn't measure how you actually respond to alcohol on a given day and it can contain read errors. Treat it as one educational data point and discuss alcohol and your health with a clinician.

If I flush, does that mean I have to stop drinking?

Not necessarily, and this post isn't here to tell you what to do. What the research supports is a reason for moderation: because acetaldehyde is a carcinogen, people who carry the inactive ALDH2 variant and drink regularly have a higher risk of upper-GI and head/neck cancers. Knowing your genotype simply lets you make a more informed choice, ideally with a clinician's input.

Does a 'normal' ADH1B and ALDH2 result mean alcohol is safe for me?

No. These two genes explain only part of how you handle alcohol. A common ADH1B plus active ALDH2 profile means you're less likely to flush, but it says nothing about your overall risk from drinking, which is influenced by dose, frequency, liver health, other genes, and lifestyle. Alcohol's health effects aren't determined by these two markers alone.

Which raw-data markers should I look for?

The two well-studied SNPs are rs1229984 for ADH1B (the fast-metabolizer variant) and rs671 for ALDH2 (the inactive variant behind the flush). Quanome reads both directly from your raw file on your device. For a step-by-step of finding rs671 in your export, see the ALDH2 flush guide.