Illustration by Gabriel Alcala

What Are Your Genes Telling You to Eat?

Companies are using your DNA to customize a diet plan. If it works, it might not be for the reasons advertised.

You’re in luck if your ideal diet plan involves a heaping helping of data. A new company called Habit will have you drink a special shake, prick your finger three times, and send the blood to a lab along with cheek swabs and detailed information about your body measurements and activity levels. A few weeks later you’ll receive dietary advice personalized to your genes and metabolism. To help you follow that advice, you can order ready-made meals.

Backed by $32 million in funding from Campbell Soup, Habit is part of a booming field of companies, with names like Orig3n and DayTwo, that promise a ultra-tailored nutritional experience. But it’s yet to be proven whether that tailoring matters.

Neil Grimmer, Habit’s founder and CEO, calls the company’s approach “test to table.” Habit’s analysis starts at $299 and looks at more than 60 biomarkers. From the cheek swabs, a lab analyzes your DNA for genetic variants that affect things like weight, metabolism, and caffeine sensitivity. The shake is similar to what you’d drink during a glucose tolerance test in a doctor’s office, but includes fat and protein too. By analyzing the blood you’ve drawn at three points in time after you’ve had the shake, Habit can study how you process those nutrients.

From the results, Habit assigns customers one of seven “habit types.” These are like personality types with a hunter-gatherer feel: “protein seekers” should embrace proteins like fish and limit carbohydrates; “plant seekers” can eat more carbs; “range seekers” have a lot of leeway. Customers also get coaching from a dietician and can get meals matched to them delivered to their door, ready to be heated up. It costs $8.99 for breakfast and $13.50 for lunch or dinner such as lamb meatballs or a mango chicken noodle bowl. “They’re nutrient dense, but they’re also data rich,” Grimmer says of the foods.

Neil Grimmer and the Habit test kit. (Courtesy of Habit)

Those who’d rather plan their own meals can follow a list of “food philosophies” for their diet type. Protein seekers, for instance, should keep vegetables on their plates and watch out for hidden fats and sugars — nothing too groundbreaking, in other words. “They’re all healthy recommendations,” says Erin Barrett, Habit’s head of scientific affairs. While every type’s food rules are widely applicable, she says, different people will have stricter or more flexible guidelines around certain food groups.

Habit launched in the Bay Area late last year and expanded to most of the country this summer. Grimmer acknowledges that the seven dietary categories are far from definitive. Habit could, he says, give “thousands of different personalized recommendations.” But by clustering these diets into the seven “habit types” Habit hopes to foster communities, he says. He thinks it will become an antidote to one-size-fits-all diet trends. “We kind of feel like we’re the fad-diet buster,” he says.

But personalized nutrition has become its own fad. Peter Jones, a professor at the University of Manitoba, estimates there are dozens of companies worldwide offering bespoke nutritional advice based on customers’ genes, microbiomes, or other factors. Scientists in the field are “getting in line to participate with these companies,” he says — himself included. Jones founded a company similar to Habit called SNPitty, still in its early stages. (The simple genetic variations that these companies usually test for are called single-nucleotide polymorphisms or SNPs, pronounced “snips.”)

Habit is “ahead of the curve in a lot of ways,” Jones says, especially with its metabolic challenge shake. SNPitty won’t use a shake, but Jones wants to analyze more SNPs than Habit does and make dietary recommendations based on how hundreds of genes might work together.

“We kind of feel like we’re the fad-diet buster,” Grimmer says. But personalized nutrition has become its own fad.

No matter how many genes you analyze, they tell you only so much. That’s why the Academy of Nutrition and Dietetics urged caution in a 2014 position paper on nutritional genomics. Conditions such as heart disease, obesity, and diabetes arise from many genetic and environmental factors working together, the authors wrote, so knowing about a single genetic variant isn’t all that useful. This field of science “holds promise,” the paper concluded, but “is not ready for routine dietetics practice.”

A 2015 meta-analysis looked at the data behind 38 genes regularly included in private companies’ nutritional genomics tests. The authors found that data for most of these genetic variants either were inconsistent or hadn’t been reproduced. “As solid scientific evidence is currently lacking,” they wrote, “commercially available nutrigenomics tests cannot be presently recommended.”

Jones acknowledges that the field is in its early days. “We all have to be cautious about getting too enthusiastic with the data we have,” he says.

But even if personalized nutrition plans don’t work in the ways they claim to, they could still help people eat better. A 2017 study randomly assigned 1,269 European adults to different diet plans. One group got advice based on their current diets; another got advice based on that information plus their body-mass index, waist circumference, and blood biomarkers; and a third set of recommendations were based on all those factors plus variations in five genes. A control group got generic diet advice. Six months later, subjects in all the personalized diet groups were eating more healthily than the control group. It didn’t matter which personalized group they’d been in — guidelines based on their starting diet were just as helpful as guidelines accounting for their physiology or genes.

So signing up for a diet based on your genes might help, and it probably won’t hurt. Except for the finger pricking.

A single golf clap? Or a long standing ovation?

By clapping more or less, you can signal to us which stories really stand out.