Health at 23andMe: The Circus of Human Traits

Editor’s note: Pending an FDA decision, 23andMe no longer offers new customers access to health reports referred to in this post. Customers who received their health information prior to November 22, 2013 will still be able to see their health reports, but those who purchased after that time will only have access to ancestry information as well as access to their uninterpreted raw data. These new customers may receive health reports in the future dependent on FDA marketing authorization. Do you have your father’s eyes or your mother’s curly hair? Do you have trouble drinking milk, or flush when you drink alcohol? Many traits such as eye color, alcohol flush, and the texture of your hair are determined—at least in part—by genetics. Learning about them also helps make genetics more real. We can observe many traits directly, compare them between people, and see how some of them are passed down through families.  But traits aren’t limited to things you can see. They include whether you can smell certain odors, the levels of hormones or other molecules circulating in your body, and even whether you’re more resistant or susceptible to infectious microbes. The circus of human traits can be wide, weird—and wildly wonderful. With 23andMe’s Trait reports, you can get a glimpse into more than 40 different human traits influenced by genetic factors. While you can “verify” certain things about yourself using these reports, it’s important to keep in mind that genetics is rarely the whole story. Your environment, lifestyle, and personal and family history can play a major role in what foods you like, your cholesterol level, whether you contract an infectious disease, and how tall you are, to name just a few things. So what do my genetics say about some of my traits?

As I mentioned in the previous post about Drug Response, I do flush when I drink alcohol. This is because I have one copy of a variant that disrupts a gene called ALDH2, which is involved in breaking down a toxic byproduct of alcohol. The disruptive variant is fairly common in East Asians, so it’s not surprising that I have it. My dad also has this variant and displays the alcohol flush trait, but my mom has two working copies of the gene. As luck would have it, my two brothers won the genetic lottery on that one—they got working copies from both parents! Another trait that displays population-specific differences is Earwax Type. Most people don’t even know there are different types of earwax, but our ancestors probably all had wet earwax, and dry earwax arose later in northeast Asia, most likely due to a single variant in a gene called ABCC11. If you’re East Asian like I am, your earwax is probably the dry type (check!), but if you’re not, you probably have wet earwax. Not all traits are determined by single genes like Alcohol Flush and Earwax Type. Many, like Eye Color and Hair Curl, are influenced by a number of genes, and 23andMe’s Trait reports may only provide information on a subset of them. Other traits can be affected by non-genetic factors; Lactose Intolerance, for example, can be partly or effectively compensated for by the types of bacteria you have in your gut (indeed, this is probably what’s happened in my case, for I used to be less tolerant of dairy products than I am now). For these reasons, the results suggested by the reports may sometimes differ from your personal experience.

A common example of this comes up with the Eye Color trait—the report may say “Likely Blue” but you might actually have brown eyes, or vice versa. The discrepancy stems from the fact that the genetic variant presented in the report is not the only variant that determines eye color. It does, however, give pretty good estimates: if you have two copies of the A version of the SNP (like I do), you have an 85% chance of having brown eyes and a 14% chance of having green eyes; if you have two copies of the G version, you have a 72% chance of having blue eyes and a 27% chance of green eyes. One copy each of A and G gives you an intermediate probabilities of brown, green, or blue eyes. So while the report may say that you are likely to have brown, or blue, eyes, there is still a chance that you don’t! As more research uncovers additional variants that influence eye color, we may be able to update the report to improve the predictions. All of our reports offer the curious individual an opportunity to learn more about the genetics—and non-genetic factors—underlying human health and traits, from the everyday to the bizarre. You might be amazed to discover that some people can smell a specific odor in their urine after eating asparagus (or can’t smell it!) or that others sneeze whenever they go out into bright sunlight, a condition charmingly referred to as “ACHOO syndrome.” Or perhaps you’re more interested in learning about how your genetic variants might influence how you respond to different diets or exercise. With traits, there’s always something for everyone!
Curious about possible traits for your child? Or wonder what your chances were of having green eyes? Customers can use the Inheritance Calculator tool to see offsprings’ probability for 6 common traits using different combinations of “parents.” (Note, you must be sharing genomes with someone to select him or her as a “parent” in Inheritance Calculator.)
Have some unique or favorite traits of your own? Customers can help contribute to research on the genetics of dozens of human traits by taking surveys about footedness, handedness, physical features, and pigmentation, among others.
More Health at 23andMe posts: Health at 23andMe: What Can You Learn?Navigating Your Health ResultsAnatomy of a 4-Star Disease Risk ReportWhat’s Your Status?– (Carrier status for Inherited Conditions)When One Size Doesn’t Fit All – (the genetics of Drug Response)
For a refresher course on genetics or help navigating the service, visit  Genetics 101 on our website, or see our  Frequently Asked Questions. Check out our companion series,  Ancestry at 23andMe, and other articles in  23andMe How-To.
  • L. Molina

    I don’t know if your report is accurate about Type 1 diabetes. I have already been diagnosed with Type 2 (for 11 years now), and I seem to have it under control my taking metmorphin (glucophage). However, I do react to the medication. Isn’t there a test for bad results taking certain kinds of medication? I haven’t taken cumadin, but my mother reacted badly to it. She too was a diabetic. And what is worse, I can’t get my physician to pay attention to this part of my symptoms. The Internet came out with information that people who took glucophage would be helped by a periodic shot of vitamin B-12. Even after showing him the articles he still refused to ok the B-12 shot. So I switched physicians. While your site does not have that kind of information, it would be very helpful if you included it in some of the information available to the public. I can’t tell you how useful that is, or has been to me, personally. I know the body decays and we can’t live forever, but I think it is possible to have various disabilities (even small ones, like described above) and still lead a productive life even in old age. Like my mom who was a diabetic, and lived to 85, never took insulin but did have heart disease. She was very active until about 6-7 months before she died.

  • carlos

    Well, my father was 50 years old when I was born and mi mother, 44. I’m the late son of eleven brothers, 6 males, 5 females. A true disparate in the Spain of the decades of 1930-50 and still today.
    My father smoked during —I believe 30 years- a media of 40 cigarettes a day, but never drank alcohol. He never was ill. One day as he heard tobacco was bad, he, from morning to evening ceased in smoking forever. He hadn’t any respiratory trouble and had an iron willpower, but only to resist. He was timid to go away and to do business. He died I think by pure waste of his body to the age of 91 years. Instead, one of my sisters which never smoked, died from a lung cancer when she was about 65 years old.
    My worry isn’t mainly about physical health. I’m not obsessed with death, obesity or the usual problems of today’s society. My preoccupation is my father was a very unsociable and with a cool mood as me and my male brothers, being his marriage a disaster, as my mother had a strong character. Curiously, my sisters are much more sociable excepting one.
    That trait together with a strong ability with mathematics —not mine- has intrigued me during much years of my life, but the case is some of my nephews have the same lack of sociability that doesn’t include mental impairment. It’s I think the most difficult thing, but although in Spain the ghost of Civil War explained some social reserve I think there’s more than that. Well, if I can let some explanation to this, that should be enough. In Spain this character is very worrying and seen as more strange, as it’s usually a country with a mood more opened than Britain, France or Germany people.
    My older brother had a retinopathy, the oftalmologists from the decade of 1950 said he should become blind. Fortunately, he’s now 78 years old and the blindness never happened. That embittered still more the hard years. When Spain progressed in the decade of 1960-70, medicine was much more better that in older times, but yes, the retinopathy existed. My father and three of my brothers had a prostate adenoma, but not cancer.

  • Dan

    Very interesting stuff!

    How many of thee ~40 traits in 23andMe were specifically tested on the genotyping chip, and how many are just the result of close associations with markers that were tested?

    In other words, how many new traits can we expect given data from the current chip?

    I’m curious if many new genetic traits will be added to the website, or if the current 40 are essentially all we will ever see with the current chip.


    • Hi Dan,

      All of your raw data, and consequently all of your genotypes reported in the Traits reports, are directly tested on the chip used to analyze your DNA. Sometimes a scientific paper will report an association between a SNP and a trait, and that SNP won’t be on the chip. In those cases, we can sometimes substitute a SNP that is on the chip, as long as the two are perfectly correlated with each other. Any reported association where this happens should mention in the accompanying description that the SNP being reported is a “proxy” or “equivalent”, but this makes up probably < 10% of the associations reported on our website. We are constantly adding new reports and we have certainly not yet exhausted the hundreds of thousands of SNPs from the previous version of the chip, let alone the hundreds of thousands of SNPs added for the newest version. There should be many many more reports in the future!