Large scale genetic and non-genetic data plus machine learning, used to provide personalized portraits of how genes and lifestyle impact weight<img loading="lazy" decoding="async" class="alignright size-medium wp-image-25105" src="https://blog-api.23andme.com/wp-content/uploads/2017/03/Genetic-Weight-report-300x297.jpg" alt="" width="300" height="297" srcset="https://blogcms.23andme.com/wp-content/uploads/2017/03/Genetic-Weight-report-300x297.jpg 300w, https://blogcms.23andme.com/wp-content/uploads/2017/03/Genetic-Weight-report-150x150.jpg 150w, https://blogcms.23andme.com/wp-content/uploads/2017/03/Genetic-Weight-report-768x760.jpg 768w, https://blogcms.23andme.com/wp-content/uploads/2017/03/Genetic-Weight-report-1024x1013.jpg 1024w, https://blogcms.23andme.com/wp-content/uploads/2017/03/Genetic-Weight-report-880x871.jpg 880w, https://blogcms.23andme.com/wp-content/uploads/2017/03/Genetic-Weight-report.jpg 1154w" sizes="auto, (max-width: 300px) 100vw, 300px" />
Have you ever wondered if it&#8217;s your genes keeping you from fitting into your jeans?
<h6> Not Just Genes</h6>
You&#8217;re not alone, and plenty of research shows that the genetic cards we are dealt at birth play an important role not just in our weight but in how much our bodies respond to diet and exercise. Our weight is determined by a complex combination of lifestyle, environment, and genes. 
 Armed with a collection of data from over 600,000 research participants, 23andMe scientists used machine learning techniques to estimate BMI based on genetics. Key results from these analyses also showed that lifestyle factors such as diet and exercise may affect weight. That affect is also dependent on a person’s genetic makeup.
<h6> What is the Genetic Weight Report</h6>
This research resulted in a new, first-of-its kind Genetic Weight report. It incorporates both genetics and lifestyle and calculates a person’s  genetic predisposition to weigh more or less than average. The report also provides insight into which lifestyle factors might make the biggest difference for a person’s own weight. 
 This report breaks new ground because it&#8217;s the first time that 23andMe scientists have been able to leverage our internal research data. This is an immense amount of information generously contributed by research participants for a health-related report. It is based on over 300 variants found to be associated with weight in 23andMe&#8217;s own database. It also customizes results based on age, sex, and ancestry.
<h6> Gene-by-Environment</h6>
23andMe scientists have also incorporated important in-house findings about lifestyle and genetics. These so-called &#8220;gene-by-environment interactions&#8221; are based on data contributed by over 45,000 research participants. 
 Each personalized report provides a ranked list of lifestyle factors found to be associated with the biggest weight differences. This is based on each person&#8217;s genetic result. In other words, this information shows how people&#8217;s specific genetics may influence their own responses to not only exercise and certain food choices, but also to factors like sleep and stress. This could help customers prioritize their own choices for maintaining a healthy weight. 
 The Genetic Weight report marks a first step towards the kind of personally tailored <a href="https://www.23andme.com/dna-health-ancestry">wellness reports.</a>  Future reports will build on the insights found through creating a machine-learning model. These type of AI models leverages the massive amount of data from customers who consent to participate in research. 
 For more on the science behind this report see our <a href="https://permalinks.23andme.com/pdf/23_17-GeneticWeight_Feb2017.pdf">white paper</a>. 
 <img loading="lazy" decoding="async" class="alignright size-large wp-image-25104" src="https://blog-api.23andme.com/wp-content/uploads/2017/03/Genetic-Weight-2-625x1024.jpg" alt="" width="625" height="1024" srcset="https://blogcms.23andme.com/wp-content/uploads/2017/03/Genetic-Weight-2-625x1024.jpg 625w, https://blogcms.23andme.com/wp-content/uploads/2017/03/Genetic-Weight-2-183x300.jpg 183w, https://blogcms.23andme.com/wp-content/uploads/2017/03/Genetic-Weight-2-537x880.jpg 537w, https://blogcms.23andme.com/wp-content/uploads/2017/03/Genetic-Weight-2.jpg 708w" sizes="auto, (max-width: 625px) 100vw, 625px" />

23andMe Blog

23andMe Blog Team

23andMe Releases First-of-its Kind Genetic Weight Report

If you think that opposites attract, you are likely mistaken, or at least that&#8217;s what the science says. Over a hundred years ago, researchers first discovered that couples are actually more alike than <img loading="lazy" decoding="async" class="alignright size-medium wp-image-17437" src="https://blog-api.23andme.com/wp-content/uploads/2013/02/AfricanAmericanCoupleHeart-200x300.jpg" alt="AfricanAmericanCoupleHeart" width="200" height="300" srcset="https://blogcms.23andme.com/wp-content/uploads/2013/02/AfricanAmericanCoupleHeart-200x300.jpg 200w, https://blogcms.23andme.com/wp-content/uploads/2013/02/AfricanAmericanCoupleHeart.jpg 283w" sizes="auto, (max-width: 200px) 100vw, 200px" />they are different.


One of the first things researchers noted was that tall people tend to partner with other tall people. Since then, studies of this phenomena – called &#8220;assortative mating&#8221; – have found strong correlations among couples for almost all traits, including behavior, personality, and disease risk.


There are  a lot of reasons for this. Love is complicated, and figuring out precisely why couples get together is tricky.
But a new study the journal <a href="https://www.nature.com/articles/s41562-016-0016">Nature Human Behavior</a> has identified genetic evidence for assortative mating, reporting that, at least for one trait, height, there is a strong correlation between the genetic variants that influence height in one partner and the actual height of the other.
In this  study, researchers from the University of Queensland in Brisbane, Australia, the University of North Carolina, 23andMe and dozens of other institutions, were able to look specifically at genetic factors by using a new design and analytical approach that allowed for them to carefully control for non-genetic influences. 
Isolating all the different cultural and social influences involved in why people partner up, isn&#8217;t easy, however. We may be attracted to our partners for any number of reasons. For instance, partners might have similar BMIs because people are attracted to individuals with a BMI that resembles their own. Or, they might choose a partner who is culturally similar. Alternatively, their BMIs might converge over time due to shared environment.


To isolate some of those kinds of influences the researchers in this study used data from more than 24,000 couples, including 11,908 from 23andMe, and compared an individual’s genetically-predicted height and BMI to those of their partner. Because genetics are determined at birth, these shouldn’t be correlated if partners become more similar only after meeting each other.
The researchers also controlled for population, to remove the effects of shared cultural background.
They found a strong correlation between people’s genetic predisposition for height and the actual height of their partner. The findings showed that similarity in couples’ heights can be largely explained by people selecting mates who are like them. Tall people, it seems, are attracted to other tall people.
But the findings were not as strong for BMI, where there was a weaker correlation between people’s genetic contributions to BMI and the actual BMI of their partners.
One possible implication of this is that “mate choice” may, in turn, affect the genetic architecture of traits in humans as well. Assortative mating may increase the odds of parents passing on a trait, like height, to their children.


“These results have implications for the interpretation of resemblance between relatives and for estimates of genetic parameters in populations,” the researchers said.
You can find the paper in the journal <a href="https://www.nature.com/articles/s41562-016-0016">Nature</a>.

Genetic evidence of assortative mating in humans

A lot of people bite their nails. It’s not that unusual, but researchers looking at the underlying biology for individuals who do it often are learning that frequent nail-biting may point to more serious conditions.<img loading="lazy" decoding="async" class="size-full wp-image-21024 alignright" src="https://blog-api.23andme.com/wp-content/uploads/2014/05/@PITER-23andMe-ChaoTian-25.jpg" alt="@PITER-23andMe-ChaoTian-25" width="199" height="287" />


For people who frequently bite their nails, the habit is associated with higher chances of having some psychiatric disorders or compulsive behaviors, according to new research by 23andMe senior statistical geneticist Chao Tian. In a short talk at the annual American Society of Human Genetics meeting in Vancouver, Chao presented findings showing   genetic associations with frequent nail-biting as well as other associations with the habit.
Before diving into her findings, it’s important to note again that in most cases nail-biting is not out of the ordinary. In children and teens, it <a href="https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3556753/">isn’t associated</a> with any psychiatric disorder. It is more a reflection of boredom, stress or even behavior modeled from parents, according to psychiatrists who study the condition.
The habit is fairly common. <a href="https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3556753/">Estimates</a> are that about 30 percent of children and 45 percent of teens bite their nails, and as people move into adulthood the percent of people who bite their nails diminishes.
But where there are interesting associations are among those who report that they frequently bite their nails. For those individuals the habit is associated with higher chances of having some psychiatric disorders or compulsive behavior, according to new research by 23andMe senior statistical geneticist Chao Tian.
In her talk at the annual American Society of Human Genetics meeting in Vancouver this week, Chao presented findings showing both genetic associations with frequent nail-biting as well as associations between the habit and certain psychiatric disorders.


For this study, Chao looked at data from 180,000 23andMe customers of European ancestry who consented to research. About 37 percent of whom reported that they bite their nails, with about seven percent of the total saying they did so frequently.<img loading="lazy" decoding="async" class="alignright size-full wp-image-24474" src="https://blog-api.23andme.com/wp-content/uploads/2016/10/Nail-biter-1.jpg" alt="nail-biter" width="147" height="193" />
Chao and other researchers at 23andMe found two dozen genetic associations with nail-biting, known by the medical term onychophagy.
While nibbling on nails during times of stress doesn’t mean one has a psychiatric disorder, among those who report that they frequently bite their nails the picture changes a bit.
Data shows that frequent nail biters are more likely to have certain disorders and habits such as depression, anxiety, and trichotillomania, the scientific name for the urge to pull out one’s own hair.
What’s also interesting is that the function of some of the genes in which nail biting associations were found point to links with things like nail structure, BMI, taste perception and neuronal regulation in the brain, Chao said.


What’s important to note here is that most cases of nail biting among children and teens isn’t associated with any psychiatric disorder, but rather a sign of boredom, stress or even behavior modeled from parents. About 30 percent of children and 45 percent of teens bite their nails. Nail-biting tends to diminish as people move into adulthood.
In this study, Chao also looked other characteristics like conscientiousness, openness, neuroticism, agreeableness and perceived stress, and found that those people who reported biting their nails often were also more likely to be neurotic and stressed. On the other hand, her research showed that people who didn’t bite their nails were more likely to be conscientious.


In and of itself this study is interesting, but it also illustrates the power of collecting large amounts of phenotypic data with genotypic data to study the underlying biology of many different kinds of conditions.

A Real Nail Biter

By looking at data on when boys’ voices first broke as a way to time the onset of puberty, researchers at <img loading="lazy" decoding="async" class="alignright size-medium wp-image-23127" src="https://blog-api.23andme.com/wp-content/uploads/2015/10/Boys-Choir-300x300.jpeg" alt="Boys Choir" width="300" height="300" srcset="https://blogcms.23andme.com/wp-content/uploads/2015/10/Boys-Choir-300x300.jpeg 300w, https://blogcms.23andme.com/wp-content/uploads/2015/10/Boys-Choir-150x150.jpeg 150w, https://blogcms.23andme.com/wp-content/uploads/2015/10/Boys-Choir-271x271.jpeg 271w, https://blogcms.23andme.com/wp-content/uploads/2015/10/Boys-Choir-372x372.jpeg 372w, https://blogcms.23andme.com/wp-content/uploads/2015/10/Boys-Choir.jpeg 445w" sizes="auto, (max-width: 300px) 100vw, 300px" />the University of Cambridge have completed a first-of-its-kind <a href="https://www.nature.com/ncomms/2015/151109/ncomms9842/full/ncomms9842.html" target="_blank" rel="noopener noreferrer">genetic study</a> and found that for both boys and girls puberty is influenced by many of the same genetic variants.
“Our study shows that although there are obvious physical differences in pubertal development between boys and girls, many of the underlying biological processes governing it are the same,” said study lead author <a href="http://www.mrc-epid.cam.ac.uk/people/john-perry/" target="_blank" rel="noopener noreferrer">Dr. John Perry,  </a> of the <a href="http://www.mrc-epid.cam.ac.uk/" target="_blank" rel="noopener noreferrer">Medical Research Council (MRC) Epidemiology Unit</a> at the University of Cambridge.


The <a href="https://www.nature.com/ncomms/2015/151109/ncomms9842/full/ncomms9842.html" target="_blank" rel="noopener noreferrer">study</a>, published in Nature Communications, is the biggest large-scale genetic study of puberty to look at both young men and women together. Part of what is unique about this study is the novel way researchers determined the age of puberty in young men.
Using data from 55,000 male 23andMe customers who consented to research, the authors of the study used the participants recollection of their age when their voice first broke as a marker for puberty and found genetic variants associated with earlier onset of puberty in boys.


“(The study) also shows that the age when men’s voices break, even when recalled decades after the event, is an informative measure of puberty timing” co-author <a href="http://www.mrc-epid.cam.ac.uk/people/felix-day/" target="_blank" rel="noopener noreferrer">Dr. Felix Day</a>, who is also at the MRC Epidemiology Unit at the University of Cambridge.


Up until now most of scientists understanding of the genetics of puberty came from studies that look at either rare disorders in young men, or studies that focused on young women alone. This novel way of looking at puberty in young men – asking men to recall the age they were at when their voices changed – allowed researchers to compare the genetic influence on the timing of puberty in both sexes.
What they found was surprising, more than a dozen genetic variants that influence the timing of puberty in women, also influence the timing of puberty in men.
In addition, the researchers also found that some of those genetic variants are correlated with conditions developed later in life such Type II diabetes, cardiovascular disease, obesity and fasting insulin levels.


The researchers picked voice breaking to gauge the onset of male puberty because it is a distinct development milestone in young men that happens when their larynx lengthens due to exposure to more male hormone.
After they looked at the genetic associations among men for onset of puberty, the researchers then compared that data to information from more than 180,000 women who were part of a consortium study, called <a href="http://www.reprogen.org/" target="_blank" rel="noopener noreferrer">ReproGen</a>, as well as from the <a href="http://www.1000genomes.org/home" target="_blank" rel="noopener noreferrer">1000 Genomes Project,</a> and data on puberty from more than 76,000 women who are also 23andMe customers and consented to research.
What they found was that of the 14 genetic variants found to be associated with puberty, five had never before been associated with puberty.
The nine other variants had been previously identified as associated with the timing of puberty in young women, but this was the first time any researchers found an association for the timing of puberty in young men.
This study is the first to quantify the strongly shared genetic basis for puberty timing between the sexes, and, according to the researchers, the results are consistent with previous studies that looked at sex-specific rare disorders of puberty.


The study also offers additional insight into the biological pathways and process of early puberty, by looking at the function of the genes and gene regions where these variants are located. For example, two of the newly found variants are located in or near the genes RORB and RXRA, which regulate cell differentiation, cell development, circadian rhythm and metabolism.
“Other studies have shown that earlier age at puberty in girls predicts higher risks for the same diseases. We now show that the same is true in men,&#8221; said <a href="http://www.mrc-epid.cam.ac.uk/people/ken-ong/" target="_blank" rel="noopener noreferrer">Dr. Ken Ong</a>, of the MRC Epidemiology Unit and a co-author of the study. &#8220;The next steps will be to understand how to prevent early puberty in boys and girls, possibly by reducing childhood overweight and obesity, or by other means.”


But he added that while many of those correlations with earlier puberty were for adverse conditions, earlier puberty was correlated with at least one favorable condition – higher bone mineral density.

Genetics found to influence puberty timing in girls and boys

By Nick Furlotte and Shirley Wu
For many of us, the battle of the bulge is more like an all-out war.
We all know people who can eat whatever they want, not work out, and yet not gain a pound. Meanwhile, eating just one burger, or missing just one cardio session, can weigh much more heavily on others (pun intended).  No doubt many of the differences we observe in weight gain and its relation to food intake and exercise are due to genetics.<img loading="lazy" decoding="async" class="alignright size-medium wp-image-19018" src="https://blog-api.23andme.com/wp-content/uploads/2013/08/Scale-300x282.jpg" alt="Scale" width="300" height="282" srcset="https://blogcms.23andme.com/wp-content/uploads/2013/08/Scale-300x282.jpg 300w, https://blogcms.23andme.com/wp-content/uploads/2013/08/Scale.jpg 357w" sizes="auto, (max-width: 300px) 100vw, 300px" />
At 23andMe we are very interested in identifying the genetic components that contribute to these differences between individuals. So we assessed people’s food and exercise habits, their physical characteristics (like whether they are “apple” or “pear” shaped), and we compared these answers to genetic data for tens of thousands of customers with European ancestry.
<h6>How do fast food and exercise affect weight on average?*</h6>
In our customers, a man of average height (5’9), average age (35) and European ancestry weighs about 180 pounds (corresponding to a body mass index (BMI) of 26.5, lower than the<img loading="lazy" decoding="async" class="alignright size-medium wp-image-19022" src="https://blog-api.23andme.com/wp-content/uploads/2013/08/WeightByGRS-female-300x195.jpg" alt="WeightByGRS-female" width="300" height="195" srcset="https://blogcms.23andme.com/wp-content/uploads/2013/08/WeightByGRS-female-300x195.jpg 300w, https://blogcms.23andme.com/wp-content/uploads/2013/08/WeightByGRS-female.jpg 631w" sizes="auto, (max-width: 300px) 100vw, 300px" />
The story is similar for the average height and average aged woman (5’4 and 35 years old) with European ancestry, who weighs about 145 pounds (a BMI of 24.8, much lower than the U.S. average of 29).  She will weigh about eight pounds less if she exercises once or twice a week, but will weigh three pounds more if she also eats fast food once or twice a week.
<h6>Why you should care if you’re apple or pear</h6>
None of this is surprising — exercise leads to lower weight on average while eating fast food leads to higher weight on average.  But it turns out that body type makes a difference.
In our customers, men with pear and apple body types weighed more if they ate fast food occasionally while men with a neutral body type weighed less with the same fast food intake. For women, having an apple shape is associated with a higher average weight (around 155 pounds) but there is very little difference in weight due to occasional fast food intake between the different body types.
Just don’t go overboard: women who ate fast food nearly every day tended to weigh about 20 pounds more than the average, and this went up by an additional six pounds for women with apple or pear shapes.
<h6>More reason to exercise</h6>
We all burn calories while we’re just sitting around, but wouldn’t it be great if you could burn more? It all depends on your “metabolic rate”, or how much energy (calories) your body uses while resting, and you can actually increase your metabolic rate by exercising regularly.
In 23andMe customers, we see this play out in the fact that physical activity level affects the degree to which fast food consumption impacts weight.  For example, an average man who eats fast food once or twice a week who also works out five to seven times a week weighs five pounds less than a man eating the same amount of fast food who does not work out at all. The story is similar for women: eating fast food and not exercising is associated with weighing 10 pounds more than average, while those who ate the same amount of fast food but also exercised frequently weighed only seven pounds more.
Of course, the difference in weight might not be due just to exercise frequency. People who exercise might also be doing other things that lead to lower weight, including eating a healthier diet overall. But the fact that exercise raises the metabolic rate makes it a plausible explanation.
<h6>Adding genetics to the picture</h6>
So how do our genetics influence all of this?    We know that certain genetic factors predispose to obesity while others may protect against it. But a recent study published in<a href="http://www.plosgenetics.org/article/info%3Adoi%2F10.1371%2Fjournal.pgen.1003607"> PLOS Genetics</a> adds a twist. The researchers showed that a set of 12 genetic factors known to be associated with obesity had less of an effect in people who exercised more and a larger effect in people who did not exercise as often.
We examined the same idea using the data from our customers and found similar results.  In women who do not exercise, the genetic risk factors were associated with weighing 1.4 pounds more than average, while women who exercised weighed only 0.75 pounds more for each risk factor.  In other words, lifestyle may actually influence the effect our DNA has on our weight.
As the size of the weight loss industry attests, weight and obesity are very challenging problems. But with more data, we’ll be able to unravel the relationship between food intake, exercise, genetics, and weight gain even more, hopefully leading to more personalized and effective healthy weight strategies.
* Analyses of average weight controlled for age, exercise frequency, fast food intake, and ancestry substructure.

Nick Furlotte

The Genetics of Bellies and Burgers

<img loading="lazy" decoding="async" class="alignright size-medium wp-image-17362" title="Woman weighing herself on scales in health club" src="https://blog-api.23andme.com/wp-content/uploads/2013/01/womenScale-300x214.jpg" alt="" width="300" height="214" />
There are lots of things we can&#8217;t change, but our lifestyle isn&#8217;t one of them. Good thing, too  – our weight, nutrition habits, and exercise behaviors impact us beyond whether we look good in a swim suit. They also influence our risk for many different diseases.
In the same way, our DNA also contributes to our risk for various diseases. It&#8217;s where our lifestyle and DNA intersect that we have the greatest potential for positive change. For instance, the knowledge that my genes put me at higher-than-average risk for heart disease may make me think twice about the extra slice of bacon (so long as it&#8217;s not within smelling distance!). And going light on the bacon for the long haul may ultimately lower my risk for many conditions besides heart disease.
So where might your genetic incentives be?
<div>Obesity</div>
Extra pounds equate to extra risk for numerous <a href="https://www.23andme.com/dna-health-ancestry/">health conditions</a>. The obese are seven times more likely to develop type 2 diabetes, three times more likely to develop gallstones, and twice as likely to develop coronary heart disease or venous thromboembolism (blood clots) compared to those at a healthy weight.
Each of these obesity-associated conditions has at least one 23andMe genetic report (see the links above).  If you find that your genetics puts you at higher-than-average risk for one of these conditions, perhaps this knowledge will motivate you to keep your weight in check.  To the fortunate few with lower-than-average genetic risk for all the conditions above, don’t grab that donut yet.  Obesity still increases your risk compared to those at a healthy weight, and there are likely many additional genetic factors that are as-of-yet unknown but may contribute to your risk.
<div style="padding: 10px; margin-left: 10px; margin-bottom: 10px; float: right; border: 1px solid #ccc; width: 280px;">
<img loading="lazy" decoding="async" class="alignleft size-full wp-image-12163" title="small_23andMe_logo" src="https://blog-api.23andme.com/wp-content/uploads/2012/04/JRBgdIBXKIeVIdETJrxYIw_ttam_logo.jpg" alt="" width="24" height="40" />
Not yet a 23andMe customer? Learn what your DNA says about your ancestry. <a href="https://www.23andme.com/store" target="_blank" rel="noopener noreferrer">Visit our store</a>!
</div>
Nutrition 
Poor nutrition, even discounting its effect on your waistline, certainly doesn’t keep the doctor away.  People who consume red meat ten times per week or more have about a 50 percent higher risk for age-related macular degeneration, an eye condition that can lead to blindness. Furthermore, studies suggest that a diet high in fiber, vegetables, nuts, and foods typically found in Mediterranean cuisine is associated with 20-40 percent lower risk for coronary heart disease. Colorectal cancer is also linked with poor diet habits, and experts suggest that  diets high in fiber and low in fat can lower your risk.
If you find that you are at increased genetic risk, this knowledge may help you prioritize which dietary changes to make. For example, a person who has a strong genetic risk for age-related macular degeneration might consider eating more fish, nuts, and leafy greens  – foods linked to lower risk for that disease.
Exercise 
A gym membership (if actually used) may positively impact your health. For starters, exercise helps keep one’s weight in check; people who exercise regularly are up to 60 percent less likely to be overweight than people who don&#8217;t exercise.
Even beyond that, exercise can strengthen one of your most important muscles: your heart. People who don&#8217;t exercise are about twice as likely to develop coronary heart disease as people who exercise regularly. In women, walking just an hour a week is associated with 50 percent lower risk for coronary heart disease and in men, vigorous activity and playing sports may lower the risk for heart disease.   If your genetics suggest you&#8217;re at higher than average risk for heart disease, take that as extra incentive to hit the gym regularly!
Poor exercise also influences your risk for type 2 diabetes.   People who engage in moderate physical activity on a regular basis or simply walk for at least two and a half hours a week at a brisk pace are about 30 percent less likely to develop type 2 diabetes compared to people who don&#8217;t exercise.   If your genetic report shows you are already at higher than average risk for type 2 diabetes, perhaps you may want to go the extra mile to make a change.
The Competitive Advantage 
Unfortunately, health isn&#8217;t an even playing field. Just as some people have genetic variants associated with lower than average risk for disease, some are born with skinny genes (and can wear skinny jeans).
<div>Still, there are other genetic factors that may help in the battle of the bulge. One that&#8217;s received a lot of attention is in a gene called PPARG. Normally, high fat intake is associated with weight gain, but in people with a certain variation of the PPARG gene, a diet high in monounsaturated fat is actually linked to weight loss. Monounsaturated fat is found in olive oil, so this type of diet has been nicknamed the &#8220;Mediterranean&#8221;-style diet.</div>
Very few genetic factors have been reliably linked to diet and exercise but some of the preliminary findings..
Whatever your health goals may be, we wish you success that outlasts the month.  May this knowledge about your DNA  – whether that be the genetics of disease risk or the genetics of losing weight  – help you towards a fitter you.

Your Lifestyle, Your Health, Your Genes

Last week we quizzed our readers on traits associated with BMI among 23andMe’s customers. Congratulations to Suellen, the winner of the quiz! She got all three answers correct and will receive an Amazon gift card. So what were the correct answers? Read on&#8230;
Which physical trait is most associated with BMI? Shoe size.
<div style="margin-top: 15px; margin-bottom: 15px; border: 1px solid #ccc; width: 600px;"><img decoding="async" class="aligncenter size-large wp-image-12686" title="BMI-physical" src="https://blog-api.23andme.com/wp-content/uploads/2012/05/BMI-physical-1024x442.png" alt="" width="600" srcset="https://blogcms.23andme.com/wp-content/uploads/2012/05/BMI-physical-1024x442.png 1024w, https://blogcms.23andme.com/wp-content/uploads/2012/05/BMI-physical-300x130.png 300w, https://blogcms.23andme.com/wp-content/uploads/2012/05/BMI-physical-768x332.png 768w, https://blogcms.23andme.com/wp-content/uploads/2012/05/BMI-physical-880x380.png 880w, https://blogcms.23andme.com/wp-content/uploads/2012/05/BMI-physical-1180x510.png 1180w" sizes="(max-width: 1024px) 100vw, 1024px" /></div>
Shoe size is very strongly correlated with BMI in 23andMe’s database. For each standard deviation increase in shoe size, BMI increases by 1.18 units. Put another way, the BMI of the bottom fifth of the shoe size distribution is 3.21 units lower on average than that of the top fifth. This corresponds to a difference of about 20 lbs in people who are 5’8” tall.
We’re really not sure what to make out of this one. Shoe size is also correlated with height but when we factored height into the analysis shoe size was still the most strongly correlated trait. Anecdotally, people’s shoe sizes can increase or decrease with weight gain and loss. It’s possible that people with wider feet may compensate by increasing their shoe size instead of choosing a wider shoe of the same size, or perhaps shoe size is a proxy for body frame, and people with larger shoe sizes have “bigger bones”. Have a theory? Add it to the comments below.
Having a round face, low arches, upturned nose, and feet that point outwards are also associated with BMI but to a lesser extent.
Which food item is most associated with BMI? Diet soda.
<div style="margin-top: 15px; margin-bottom: 15px; border: 1px solid #ccc; width: 600px;"><img decoding="async" class="aligncenter size-large wp-image-12687" title="BMI-food" src="https://blog-api.23andme.com/wp-content/uploads/2012/05/BMI-food-1024x430.png" alt="" width="600" /></div>
The surprise for us here was that ice cream consumption was not really correlated with BMI at all. On the other hand, people who drink diet soda five or more times a day have an average BMI nearly 5 units higher than those who never drink diet soda. Not to imply that we should all trade in that diet coke for a double hot fudge sundae &#8212; we don’t know whether drinking soda is a cause, a consequence, or just linked to something else that causes weight gain. These findings do, however, parallel that of a study conducted in Texas, in which normal weight individuals were followed for 8 years. People who reported drinking more diet soda gained more weight over those 8 years, suggesting that diet soda might actually lead more directly to weight gain.(1) But for now, the jury’s still out.
Some other foods were associated with BMI, too, including red meat, yogurt, and vegetables.
Which personality or behavior trait is most associated with BMI? Activity and depression.
<div style="margin-top: 15px; margin-bottom: 15px; border: 1px solid #ccc; width: 600px;"><img decoding="async" class="aligncenter size-large wp-image-12688" title="BMI-personality" src="https://blog-api.23andme.com/wp-content/uploads/2012/05/BMI-personality-1024x549.png" alt="" width="600" srcset="https://blogcms.23andme.com/wp-content/uploads/2012/05/BMI-personality-1024x549.png 1024w, https://blogcms.23andme.com/wp-content/uploads/2012/05/BMI-personality-300x161.png 300w, https://blogcms.23andme.com/wp-content/uploads/2012/05/BMI-personality-768x412.png 768w, https://blogcms.23andme.com/wp-content/uploads/2012/05/BMI-personality-880x472.png 880w, https://blogcms.23andme.com/wp-content/uploads/2012/05/BMI-personality-1180x633.png 1180w" sizes="(max-width: 1024px) 100vw, 1024px" /></div>
These are less of a surprise. Activity is strongly correlated with BMI, and is no doubt due to the calorie burning benefits of exercise. The reverse may also be true &#8212; being overweight may itself decrease activity levels, creating a vicious cycle.
For depression, the arrow of causation seems to point in both directions. For example, a recent meta-analysis found that being overweight or obese increased the risk of developing depression, and having depression increased the risk of becoming obese.(2)
Having greater levels of self-discipline was correlated with lower BMI to a lesser extent, but aside from this we didn’t find a significant relationship between BMI and any other personality factors.
Check back soon for our next quiz and another chance to win!
(1)  Fowler, SP, Williams, K, Resendez, RG, Hunt, KJ, Hazuda, HP, &amp; Stern, MP. (2008). Fueling the Obesity Epidemic? Artificially Sweetened Beverage Use and Long-term Weight Gain. Epidemiology, 16(8): 1894-1900.
(2)  Luppino, FS, de Wit, LM, Bouvy, PF, Stijnen, T, Cuijpers, P, Penninx, BWJH, Zitman, FG (2010). Overweight, Obesity, and Depression: A systematic review and meta-analysis of longitudinal studies. Arch Gen Psychiatry. 2010;67(3):220-229.

Quiz Answers: Traits Associated With BMI

<img decoding="async" class="alignright size-full wp-image-219" title="feetscale" src="https://blog-api.23andme.com/wp-content/uploads/2008/05/feetscale.jpg" alt="" width="300" srcset="https://blogcms.23andme.com/wp-content/uploads/2008/05/feetscale.jpg 425w, https://blogcms.23andme.com/wp-content/uploads/2008/05/feetscale-300x199.jpg 300w" sizes="(max-width: 425px) 100vw, 425px" />
Besides height and weight, can you guess what feature is most strongly correlated with body mass index (BMI)? Over the next few months, we’ll be testing your knowledge of some of the most interesting associations we’re seeing in the 23andMe database. Guess right, and you could win cool prizes, including gift cards and 23andMe schwag! Then come back the next week to learn the answers and see who won.
This quiz is now closed. Read on to learn more about how we find these associations! 
 
What can we discover?
23andMe helps its customers connect with the latest research on correlations between genetic variations and health conditions and traits. These genetic variations could be pieces of the puzzle explaining why some people develop, say, type 2 diabetes, or why some people have curly hair.
With the help of our customers we’ve also made new discoveries of genetic associations for things like smelling the asparagus when you pee, Parkinson’s disease, and hypothyroidism. But you can explore correlations between practically anything, and 23andMe’s database, which so far has more than 150,000 genotyped customers contributing more than 40 million answers on hundreds of different topics, is a uniquely powerful platform for discovery.
What’s in a correlation?
<div style="border: 1px solid #ccc; width: 315px; padding: 10px; margin-left: 15px; background-color: #eee; font-size: 11px; float: right;">
Does correlation imply causation?
No. Correlation sometimes points to causes &#8212; cigarette smoke contributing directly to lung cancer is a common example &#8212; but it can often be due to some other underlying factor. For example, owning a house may be correlated with lower risk of type 2 diabetes*, but this doesn’t mean that home ownership protects you from having the disease. Instead, it could be that some other factor, such as higher socioeconomic status, makes it more likely that someone is able to buy a house as well as eat a healthier diet, which helps lower one’s risk for type 2 diabetes.
To avoid these sorts of “phantom” associations, researchers often add include demographic or other factors in their analyses to see if those factors explain away the findings. This is usually referred to as “controlling for” various factors.
* fictitious example
</div>
When we search for correlations, we’re looking to see whether there’s a relationship between two different variables, such as smoking and lung cancer. Sometimes we might investigate the relationship between two specific things, but often researchers will look at many different variables to see if any of them are correlated with the outcome of interest. In genetic associations, most studies routinely look at thousands or millions of genetic variants to find correlations with a trait or disease.
When we do this, we are essentially testing many different hypotheses at once. Each variable that you investigate gives you another chance to find a correlation, but if you find one, is it because it’s a real correlation, or simply because you looked at so many variables that one of them just happened to work? We can calculate statistics to measure how likely it is for a finding to be due to chance (also called “statistical significance”), and adjust these statistics based on how many different things we’re testing. After all that’s said and done, the less likely it is that a finding is due to chance, the more likely it is to represent a real association.
More than genetics
Because 23andMe looks at numerous traits as well as genetics, we can explore not just whether genetic factors are correlated with traits, but whether different traits are correlated with each other. For example, we’ve tested BMI against approximately 1000 different traits, habits, and conditions reported by our customers. What we found might surprise you&#8230;
<a title="Quiz Answers: Traits Associated With BMI" href="https://blog.23andme.com/23andme-research/bmi-quiz-answers/">Check out this post to see what we discovered!</a>

What Do Diet and Personality Have to Do With BMI?

<img decoding="async" class="alignright size-full wp-image-219" title="feetscale" src="https://blog-api.23andme.com/wp-content/uploads/2008/05/feetscale.jpg" alt="" width="350" srcset="https://blogcms.23andme.com/wp-content/uploads/2008/05/feetscale.jpg 425w, https://blogcms.23andme.com/wp-content/uploads/2008/05/feetscale-300x199.jpg 300w" sizes="(max-width: 425px) 100vw, 425px" />
Obesity is an enormous public health concern in the United States, affecting about 33% of adults and 20% of children and adolescents.
<h2>The Genetic Contribution to Obesity</h2>
While lifestyle and environmental factors play an undeniable role in whether an individual becomes obese, researchers have also uncovered common genetic variants that influence risk for the condition. One of the best-studied variants is in the FTO gene, associated with obesity in young adults. However, studies in different age groups have revealed subtle differences in the effect of the FTO variant on obesity risk.
To better characterize the effects of throughout childhood development, an international team of researchers from the Early Growth Genetics Consortium (EGGC) analyzed data from eight study populations, totaling nearly 20,000 individuals of European descent. Their results, published last week in <a href="http://www.plosgenetics.org/article/info:doi/10.1371/journal.pgen.1001307" target="_blank" rel="noopener">PLoS Genetics</a>, showed a distinct genetic association with BMI that depended on age.
<h2>Looking at Age</h2>
In older children (ages 6-13), the variant was associated with a slight increase in BMI. The opposite was true in children younger than two years.
<img loading="lazy" decoding="async" class="alignright wp-image-8507 size-medium" title="journal.pgen.1001307.g002" src="http://35.88.189.111/wp-content/uploads/2011/02/journal.pgen_.1001307.g002-300x295.png" alt="BMI versus age by FTO genotype" width="300" height="295" srcset="https://blogcms.23andme.com/wp-content/uploads/2011/02/journal.pgen_.1001307.g002-300x295.png 300w, https://blogcms.23andme.com/wp-content/uploads/2011/02/journal.pgen_.1001307.g002.png 600w" sizes="auto, (max-width: 300px) 100vw, 300px" />
The switch in association from lower BMI to higher BMI occurred between 4-6 years of age, corresponding with a phenomenon known as “adiposity rebound” (AR). During normal development, infants experience rapidly increasing BMI, peaking around age one, followed by decreasing BMI that bottoms out around age five before rising again. AR refers to this second rise in BMI, and the EGGC study found that the A version of was also associated with earlier AR. (See figure to the right from the EGGC paper.)
<h2>FTO and Obesity</h2>
It turns out that the timing of AR can have implications later in life. <a href="https://www.nature.com/ijo/journal/v30/n4s/full/0803514a.html" target="_blank" rel="noopener">Multiple studies</a> have linked earlier AR to higher BMI in adulthood, along with increased risk for obesity-related diseases such as heart disease and type 2 diabetes.
Based on their findings, the EGGC researchers suggest that the connection between the FTO gene and obesity may have less to do with a general predisposition to obesity and more to do with childhood development and timing of AR. This observation may provide greater insight into the biological mechanisms underlying obesity, though additional studies are needed to confirm and refine the EGGC results.

BMI — Page 2