If you've ever had a friend that's dropped an amazing amount of weight on a diet, but was a miserable failure for you, it probably wasn't your fault: The fault was in your genes.

The success of different diets — whether high-fat, Mediterranean, or any other popular diet — depends on your genetic composition, which may be completely different from your friend.

"There is an overgeneralization of health benefits or risks tied to certain diets," said William Barrington, Ph.D., who conducted the study at Texas A&M University. "Our study showed that the impact of the diet is likely dependent on the genetic composition of the individual eating the diet, meaning that different individuals have different optimal diets."

The study's findings may have implications beyond just which diet works best for each individual. They could also affect dietary recommendations issued by the Food and Drug Administration, since recommendations are based on average responses of many people and may not be applicable to some individuals.

Researchers used mice for the current study. "Mice provide a powerful model for studying the effects of diets in different genetic backgrounds because they have similar susceptibilities to obesity and metabolic syndrome," said Barrington. "We can model the genetic diversity that is seen in humans while controlling for environmental factors."

The researchers used four mouse strains to represent the genetic diversity found in humans. The genetic differences between any two strains were similar to that of two unrelated people.

For six months the mice received food equivalent to today's Western diet, a traditional Japanese diet, a traditional Mediterranean diet, or a high-fat, low-carb Atkin's-like diet known as ketogenic, while some mice received standard mouse chow for comparison. The mice could eat as much food as they wanted, but the researchers kept tabs on how much was consumed.

The researchers matched the test diets closely with what people would eat on the same diet. For example, the Japanese diet used rice as the main carbohydrate and included green tea extract to mimic the effects of this bioactive compound. For the Mediterranean diet, wheat was the main carbohydrate, and red wine extract was included to imitate this key dietary component.

Mice eating the Western diet generally showed negative health effects, including increased obesity, fatty liver disease, and detrimental effects on cholesterol, but the severity of the effects varied widely depending on the strain. In fact, one strain of mice appeared largely resistant to any negative health effects from this diet.

The Western diet and the ketogenic diet, which are both high in fat, showed opposite responses for two strains of mice. For one strain, the researchers observed very negative health effects on the Western diet, including increased obesity and fatty liver disease, but saw no negative health effects when this strain ate the high-fat, low-carb ketogenic diet.

On the other hand, a different strain of mice had increased obesity and signs of metabolic syndrome on the ketogenic diet but was much healthier on the Western diet.

"We also found that the causes for obesity were different," said Barrington. "Some mice on specific diets simply ate more calories, and this caused them to become obese. However, mice on other diets ate less but still became obese."

For all the mouse strains, the ketogenic diet increased the number of calories burned without any increase in activity, but some strains of mice ate so much on this diet that they still became obese and experienced negative health effects.

"It is highly likely that the level of diversity of diet response seen in our study will also be observed in humans," said Barrington. "Since there are different optimal diets for different individuals, this underscores the need for precision nutrition, which would identify optimal dietary patterns for each person."

The researchers are now working to identify the genes and biological mechanisms involved in the varying responses to diets. This line of research could eventually lead to a genetic test that identifies who is likely to benefit or experience negative health effects from certain diets.

"We've largely viewed diet the same way for the last 100 years — assuming that there is one optimal diet," said Barrington. "Now that we've identified that this is likely not the case, I think that in the future we will be able to identify the genetic factors involved in the varying responses to diet and use those to predict diet response in humans."