In the era of biohacking, big data, and a worsening health crisis, startups are clamoring to personalize nutrition plans according to individual genetic profiles. It seems a perfect solution to the conflicting nutrition advice provided by today’s diet zeitgeist: mediterranean vs. ketogenic, paleo vs. plant-based, DASH Diet vs. Whole30. Which macronutrient is most evil? Which of these dietary equations will solve our personal and public health problems? The new answer, according to cutting edge companies like Habit, Nutrigenomix and DNAFit, is that there is no one-size-fits-all diet. Our metabolic responses differ according to our genes.

A slew of companies are now offering nutrition advice tailor-made to fit each customer’s DNA, exercise regime, lifestyle, blood markers and more. Prick your finger, swab your cheek, and send in your home test kit. They crunch the data and offer personalized nutritional insights to help you succeed where the food pyramid failed. These insights are served up alongside supports such as tracking apps, nutritionist consultation, and—in some cases—meal delivery.

The marketing is sleek and the concept is rational; but the scientific evidence that commercially available personalized nutrition plans are beneficial is still a little underripe. At the end of February, one of the largest, most rigorous nutrigenomic studies to date made headlines when it failed to show an association between certain gene variants and patients’ response to both whole-food low fat and whole-food low carb diets. There was no significant difference in outcomes between participants on low carb diets and those on low fat diets. Furthermore, participants’ genotypes had no impact on how they responded to the two dietary interventions.

These results are frustrating for researchers trying to crack the chromosomal code. As the study’s lead researcher Christopher Gardner says, "When I saw the results, this wave of disappointment washed over me. It was like, wait, it didn’t work? None of the genetic variants had an effect?”

Others have reported similarly underwhelming findings. For instance, studies by Food4Me, a European Union funded project, did not show clinically relevant results for a DNA-driven nutrition intervention. A 2015 meta-analysis found no meaningful association between genetic markers that are commonly analyzed in commercially available nutrigenomic tests and diet-driven diseases.

Why is it so hard to pinpoint causal genetic links? A multitude of variables factor into the equation: metabolic responses aren’t mediated by a single gene or even simply by a combination of genes; our bodies’ reactions hinge on genetic, microbial, and environmental interactions. It’s the old nature versus nurture question, and focusing exclusively on genetic markers is ignoring a substantial piece of the equation.

DNAFit, which uses 38 genes to customize diet and fitness advice, offers consumers the caveat: “Genetics is only one part of the picture, so we never overstate the application of genetics in the real world. The value comes from understanding your genetic profile in conjunction with your goals, lifestyle and environment—the whole picture.”

Nutrigenomix, which similarly uses 45 genetic markers to create personalized advice, does not work directly with consumers, but instead offers custom nutrition council through healthcare professionals who can help interpret and apply the insights.

Meanwhile, Habit takes a systems biology approach, collecting genetic information, blood samples, and self reported biomarkers (such as height and weight). Habit grinds all of these biometrics through its proprietary algorithm to produce individualized, actionable insights.

Many researchers working to establish genetic variants with dietary implications are confident that nutrigenomics holds promise—but that we’re still a ways away from delivering on that promise. “I think companies offering personalized dietary advice are probably running ahead of the evidence,” says Dr. John Mathers. Mathers is the director of the Human Nutrition Research Center at Newcastle University in Britain and lead researcher in the Food4Me study. What we currently have is a keyhole view into a vast landscape of interplaying variables, which include everything from DNA to the microbiome to the environment.

In the meantime, however, it’s worth remembering that simply eating whole foods helped the participants in Christopher Garner’s study improve their health, regardless of genotype. And it’s also worth noting that, according to the CDC’s 2017 report, only 1 in 10 adults meets the recommended dietary intake for fruits and vegetables. While scientists continue to noodle on the role of DNA in diet, most people would be well served by the unglamorous universal recommendation to eat more whole foods.