What TAS2R38 Means in Your 23andMe Raw Data

TAS2R38 is the gene behind a bitter taste receptor on your tongue. The three common TAS2R38 SNPs used to infer PAV/AVI are rs713598, rs1726866 and rs10246939. The pattern of amino acids encoded by those markers tells you whether you carry the PAV or AVI form, which changes how strongly you taste PTC, PROP, and bitter compounds in brassica vegetables.

What does TAS2R38 mean in my 23andMe raw data?

When a raw file lists TAS2R38, you are not seeing one result. You are seeing individual markers in the gene. The three common markers used to infer PAV/AVI are:

• rs713598
• rs1726866
• rs10246939

Each marker changes one amino acid in the receptor protein. The combination of the three is called a haplotype. The two common haplotypes are PAV and AVI, named after the amino acids proline-alanine-valine and alanine-valine-isoleucine. PAV builds a receptor that binds certain bitter compounds strongly. AVI builds one that binds them weakly. Rare recombination or unexpected haplotypes mean the PAV/AVI call is an inference from the three markers, not a guaranteed phenotype.

A raw line from a file can look like this:

rs713598 7 141672604 CC

Allele letters vary by build and strand, so a single marker cannot assign the haplotype. You still need all three markers to infer the full haplotype, because the three markers can occasionally mix. In practice many tools condense them into a single call: PAV/PAV, PAV/AVI or AVI/AVI.

TAS2R38 does not tell you what foods you will like. It tells you how strongly a specific set of bitter compounds will hit your tongue. Liking broccoli is still a decision, not a gene.

The mechanism: why Brussels sprouts taste different to different people

TAS2R38 makes a receptor protein in taste cells. The receptor responds to the test chemicals PTC and PROP and to bitter compounds found in brassica vegetables. When the receptor binds one of these compounds, it sends a bitter signal. The PAV version has a binding pocket that fits these molecules well. The AVI version has small shape changes that make the fit weaker, so the same Brussels sprout sends a much smaller bitter signal.

This is not a subtle preference. It changes the intensity at the same dose. People with two PAV copies often describe kale, Brussels sprouts, and broccoli rabe as sharply bitter. People with two AVI copies often call those same items mild. One copy sits in between. The gene does not make you dislike bitterness. It changes the volume of a specific bitter channel.

What your raw data actually shows

The table above shows the usual amino acid pattern across the three markers. Open the raw file, search for the rsID, and look at the last column. A file from a different company or one using the reverse strand may report different allele letters, so a tool's final PAV or AVI call matters more than any single letter.

What you will not see is a tidy line that says TAS2R38: taster. Raw files are not interpreted. That step is done by a report or a tool. It is why the same file can look like noise in a text editor and obvious in a report.

What the result does and does not say about food preferences

This is where the simple answer fails. TAS2R38 does not say you hate broccoli. It says the bitterness is louder for you. Liking a food also depends on texture, sweetness, salt, fat, smell, childhood exposure and cooking method. Cooking method can change perceived bitterness. Roasting with oil and salt may add flavours that compete with bitterness, and boiling may change texture in ways that matter more to a PAV/PAV taster.

The result also does not mean you are a picky eater. Pickiness is behavioural. It does not mean you should avoid brassica vegetables. The bitterness is not harmful. It does not even mean you are a supertaster. Supertaster is a broader label that includes other taste genes and the density of taste buds.

Practical reads:

• PAV/PAV: brassica bitterness is real. Fat, salt, acid or char changes it more than willpower does. Start with milder options like broccolini, cauliflower or roasted cabbage if kale is too much.
• PAV/AVI: bitterness is noticeable but usually manageable with normal cooking.
• AVI/AVI: bitterness is probably not the main reason you avoid these foods. Texture, smell or old associations deserve more attention.

If I am AVI/AVI, can I still taste bitter things?

Yes. Humans have roughly two dozen TAS2R genes. TAS2R38 is one of them. It handles a narrow set of compounds found in brassica vegetables and related test chemicals. Other receptors detect caffeine, quinine, many medicines and other plant toxins. So AVI/AVI does not mean you cannot taste bitterness. It means you are less sensitive to one specific bitter channel. You can be a PTC non-taster and still find black coffee bitter.

Where the evidence thins out

The effect on brassica bitterness is strong in taste tests. Evidence linking TAS2R38 to actual brassica intake is mixed, and any effect appears smaller than the effect on perceived bitterness. The gene also does not predict health outcomes by itself. Eating vegetables has benefits that do not depend on whether you enjoy them, and less bitter vegetables can provide similar fibre and nutrients.

healthosx.com/dna

The download steps for 23andMe, on the page itself.

healthosx.com/dna

The short version: what is worth acting on, first.

If you want to see your own TAS2R38 call, the free tool at HealthOS DNA reads raw files in the browser, including major consumer DNA files and whole-genome VCFs, and reports the TAS2R38 headline as "Why broccoli tastes different to you than to other people." VCFs without rsIDs get ancestry only, not the trait report. The trait report does not upload your file.