Free tool · Obesity genetics
As heritable as height.
Heritability means how much of the difference between people comes down to genes. Here it is for obesity, next to two things nobody blames you for.
How much of it is genes
Twin studies. Tap a row.
Obesity · measured in 140,379 pairs of twins
Nobody picks the size of their body any more than they pick their height. Both are set largely by the genes you were born with.
How many genes they have found
97
spots in our DNA
The first big search looked at 339,224 people and found 97 spots in our DNA linked to body weight.
There is no single obesity gene. There are thousands of them, and each one on its own is tiny. Every one of these studies was done in people, not animals, and they needed hundreds of thousands of people at a time because the effects are so small.
Why does it say spots rather than genes? Think of your DNA as a neighborhood. It runs street after street, and the genes are the houses along them. Every house has an address.
A search like this does not knock on doors. It compares hundreds of thousands of people, narrows down which addresses the difference seems to be coming from, and writes those addresses on a list. That is what a spot is. An address, not a name.
So 941 spots means 941 addresses. Somebody still has to go to each one and work out which house is actually responsible, which is much slower work. Of those 941, 138 have had a house named so far. The rest are addresses with nobody's name on them yet.
GLP-1 Strong
Facts are useful when you have to speak up for yourself
Weight stigma turns up in exam rooms and at family dinners. Knowing what the evidence actually says gives you something specific to say back.
The STRONG framework is a step by step way to respond to it while it is happening: Stop, Think, Respond, Offer Clarity, Navigate, and Grow stronger together. Offering clarity is the step where facts like the ones on this page get used.
FAQ
Is obesity genetic?
Largely, yes. Across 140,379 pairs of twins in 20 countries, between 57 and 77 percent of the difference in body weight from one person to the next came down to genes. It is highest in your twenties and eases off with age as your own life piles up.
That is the same kind of number you get for height, measured by the same researchers in the same twins. It is also about the same as type 2 diabetes, at 72 percent.
Is there an obesity gene?
No. There is no single gene that makes someone gain weight, and no test that finds one for most people. There are thousands of small ones instead, and each of them on its own shifts your weight by a tiny amount.
A very small number of people do have obesity caused by a single gene. It is rare, it usually starts in early childhood with extreme hunger, and it is diagnosed by a clinical genetics service rather than a consumer DNA kit.
How many genes are linked to obesity?
The count has gone up every time the studies got bigger. A 2015 search across 339,224 people found 97 places on the DNA linked to body weight. By 2018, with about 700,000 people, it was 941. Modern genetic scores go further again and add up more than two million tiny differences at once.
Those are places on the DNA rather than named genes. Of the 941 found in 2018, researchers have been able to name an actual gene at 138 of them so far.
Does having obesity genes mean you cannot lose weight?
No, and this has been tested directly rather than guessed at. Two large trials, the Diabetes Prevention Program and Look AHEAD, followed 1,824 and 3,906 people for four years and checked 91 of these obesity genes against how much weight each person actually lost and then kept off.
Almost none of those genes predicted anything. The researchers reported that most obesity-linked gene variants were not associated with weight loss or with regain, either on their own or through the programme people were on.
So your genes have a great deal to say about where you started. On the evidence so far, they have very little to say about how far you get.
Can a DNA test tell you if you will have obesity?
Not usefully, no. The 941 known variants together explain only about 6 percent of the differences in body weight between people, and a score built from all of them lines up with what people actually weigh at a correlation of about 0.22. That is a long way from a prediction about any one person.
There is a second problem. Almost all of this research was done in people of European ancestry, and these scores are known to be several times less accurate for everyone else. The field has said so itself.
Twin studies can see the whole inherited contribution at once without naming a single gene. A DNA chip currently reads back a small slice of it. The gap between those two is a gap in the tools, not evidence that the inheritance is not real.
If obesity is genetic, why has obesity increased so much?
Because both things are true at once, and the same twin study shows it. Between the 1940s and the 2000s, average body weight rose sharply and the spread between people widened, while the heritability figure barely moved. It also came out much the same in North America, Europe and East Asia, despite very different rates of obesity.
Genes did not change over sixty years. The surroundings did. What appears to be inherited is largely how strongly a person responds to the surroundings everyone is sharing.
Do diet and exercise still matter if obesity is genetic?
Yes. One of the clearest findings on the biggest known obesity variant is that being physically active blunts it. Across 45 studies and 218,166 adults, the FTO risk variant raised the odds of obesity by 1.30 per copy in inactive people and 1.22 per copy in active people, which the authors describe as the effect being reduced by 27 percent.
Note what that does and does not say. The effect shrinks by roughly a quarter. It does not vanish. Two people doing exactly the same things still do not end up in the same body, and that is the part the blame usually gets aimed at.
Sources
10 references
Every figure above was read from the study that reported it. One thing worth knowing: almost all of this research was done on people of European ancestry, and genetic scores are known to work less well for everyone else. The researchers say so themselves.
- Silventoinen K, Jelenkovic A, Sund R, et al. Differences in genetic and environmental variation in adult BMI by sex, age, time period, and region: an individual-based pooled analysis of 40 twin cohorts. The American Journal of Clinical Nutrition. 2017;106(2):457–466. doi:10.3945/ajcn.117.153643. The obesity bar. Heritability of BMI was 0.77 (95% CI 0.77, 0.78) in men and 0.75 (95% CI 0.74, 0.75) in women aged 20–29, falling to 0.57 and 0.59 in the oldest groups, across 140,379 complete twin pairs in 20 countries. Also the source for heritability being much the same across regions and across the decades from the 1940s to the 2000s, despite mean BMI rising sharply over that time.
- Jelenkovic A, Hur YM, Sund R, et al. Genetic and environmental influences on adult human height across birth cohorts from 1886 to 1994. eLife. 2016;5:e20320. doi:10.7554/eLife.20320. The height bar: 0.69–0.84 in men and 0.53–0.78 in women across 143,390 complete twin pairs. This is the same consortium as source 1, CODATwins, modelling both traits the same way in largely the same cohorts, which is what makes the height comparison a like-for-like one rather than two numbers from two unrelated papers.
- Willemsen G, Ward KJ, Bell CG, et al. The concordance and heritability of type 2 diabetes in 34,166 twin pairs from international twin registers: the discordant twin (DISCOTWIN) consortium. Twin Research and Human Genetics. 2015;18(6):762–771. doi:10.1017/thg.2015.83. The type 2 diabetes bar: 72 percent (95% CI 61–78) across 34,166 same-sex twin pairs aged 45 and over.
- Elks CE, den Hoed M, Zhao JH, et al. Variability in the heritability of body mass index: a systematic review and meta-regression. Frontiers in Endocrinology. 2012;3:29. doi:10.3389/fendo.2012.00029. Why the chart uses twin studies only, and why its obesity figure is higher than the 40 to 70 percent often quoted. Across 88 twin estimates in 140,525 twins, BMI heritability ranged 0.47 to 0.90, against 0.24 to 0.81 from 27 family studies. Averaging the two designs together produces a number that matches neither.
- Locke AE, Kahali B, Berndt SI, et al. Genetic studies of body mass index yield new insights for obesity biology. Nature. 2015;518(7538):197–206. doi:10.1038/nature14177. The 97 loci, found in up to 339,224 individuals.
- Yengo L, Sidorenko J, Kemper KE, et al. Meta-analysis of genome-wide association studies for height and body mass index in ~700,000 individuals of European ancestry. Human Molecular Genetics. 2018;27(20):3641–3649. doi:10.1093/hmg/ddy271. The 941 near-independent variants associated with BMI, which together explain about 6 percent of the variation in an independent sample, the polygenic score built from them correlating about 0.22 with actual BMI, and 138 genes prioritized by combining the association data with gene expression data.
- Khera AV, Chaffin M, Wade KH, et al. Polygenic prediction of weight and obesity trajectories from birth to adulthood. Cell. 2019;177(3):587–596.e9. doi:10.1016/j.cell.2019.03.028. The score built from 2.1 million common variants and tested in more than 300,000 people, which is the figure behind the third step of the gene count.
- Papandonatos GD, Pan Q, Pajewski NM, et al. Genetic predisposition to weight loss and regain with lifestyle intervention: analyses from the Diabetes Prevention Program and the Look AHEAD randomized controlled trials. Diabetes. 2015;64(12):4312–4321. doi:10.2337/db15-0441. The answer to whether these genes mean you are stuck. 91 established obesity-predisposing loci tested against weight loss across four years and against regain, in 1,824 Diabetes Prevention Program and 3,906 Look AHEAD participants. Most were not associated with weight loss or regain, directly or through any interaction with the lifestyle programme. One variant, MTIF3 rs1885988, was associated with greater weight loss under the intervention in both trials at a nominal P = 4.3 × 10⁻³, which did not reach the study-wise threshold of P < 5.8 × 10⁻⁴.
- Kilpeläinen TO, Qi L, Brage S, et al. Physical activity attenuates the influence of FTO variants on obesity risk: a meta-analysis of 218,166 adults and 19,268 children. PLoS Medicine. 2011;8(11):e1001116. doi:10.1371/journal.pmed.1001116. The FTO variant raised the odds of obesity by 1.22 per copy (95% CI 1.19–1.25) in physically active adults against 1.30 per copy (95% CI 1.24–1.36) in inactive adults, which the authors report as the association being attenuated by 27 percent. Pooled from 45 adult studies. No such interaction was found in the 19,268 children and adolescents.
- Martin AR, Kanai M, Kamatani Y, Okada Y, Neale BM, Daly MJ. Clinical use of current polygenic risk scores may exacerbate health disparities. Nature Genetics. 2019;51(4):584–591. doi:10.1038/s41588-019-0379-x. The source for the ancestry note above: polygenic scores available today are several times more accurate in people of European ancestry than in other ancestries, a consequence of bias in the underlying studies, and using them clinically as they stand would systematically favour European-descent populations.
Coaching
You are not working against a character flaw.
Obesity is a disease and it is not your fault. If you have spent years being told it was a discipline problem, knowing the biology changes what you are actually working on.
Work with me →